A New Perspective On AI Safety Through Control Theory Methodologies

Alois C. Knoll; Knut Graichen; Lars Ullrich; Mohan Trivedi; Ross Greer; Walter Zimmer

REVIEW 3 major objections 5 minor 273 references

A New Perspective On AI Safety Through Control Theory Methodologies

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper argues that AI safety can be grounded in a control-theoretic paradigm called data control, which classifies AI systems, defines robustness, sensitivity, and stability properties, and formalizes lifecycle enhancement mechanisms.

desk verdict A wide-ranging perspective with a genuinely new conceptual frame, but the central AID stability property is not a valid stability criterion and needs rework before it can ground safety claims. read the letter →

arxiv 2506.23703 v1 pith:KX3UX37C submitted 2025-06-30 cs.AI

Lars Ullrich , Walter Zimmer , Ross Greer , Knut Graichen , Alois C. Knoll , Mohan Trivedi This is my paper

classification cs.AI

keywords AIsafetycontroltheorydatasystemanalysisclassesAIDstabilitycircumstancerobustnesssensitivity

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that the safety of modern AI systems can be analyzed and assured the way control engineers analyze dynamical systems. It introduces 'data control' as a paradigm: instead of applying AI to control, apply control-theoretic system analysis to AI. The core move is to treat an AI system's input-output behavior as a statistical conditional distribution $P(Y|X)$ that evolves over time under causal, latent, contextual, and environmental influences. On that basis the paper defines three behavioral classes of AI systems—static, non-stationary, and dynamic—and three properties—circumstance robustness, circumstance sensitivity, and dynamics stability—that are meant to hold across architectures and applications. If correct, this gives developers, auditors, and regulators a common, application-agnostic vocabulary for specifying safe behavior over the whole AI lifecycle.

What carries the argument

The central object is the conditional input-output distribution $P(Y|X)$, treated as an observable probabilistic state of the AI system. The paper's machinery has three parts: the Principle of Statistical Dynamics gives this distribution an equation of motion; the class definitions (static, non-stationary, dynamic) determine which part of that motion is due to memory versus current input; and the properties—especially AID stability, which uses the KL divergence $D_{\mathrm{KL}}(P(Y|X)_t \,||\, P(Y|X)_{t-1})$ as a discrete-time Lyapunov function—turn the dynamics into checkable safety criteria.

What would settle it

Construct a recurrent network with two distinct hidden histories that produce the same marginal output distribution at time $t$, then show that a sequence with non-increasing KL divergence between consecutive output distributions can be followed by a sharp rise in prediction error or unsafe action. Such a case would show that $P(Y|X)$ is not a state variable and that the AID stability criterion can certify a system that is not actually stable.

Watch

Extended reading notes

Core claim

The paper's central claim is that an AI system can be treated as a data-based signal-processing system whose safety-relevant behavior is captured by the conditional input-output distribution $P(Y|X)$. It postulates a Principle of Statistical Dynamics, $\frac{dP(Y|X)}{dt} = F(\cdot)$, asserting that this distribution itself evolves in time due to causes, latent variables, confounders, interventions, context, task, environment, and time. From this, the paper derives a tripartite classification: static AI systems keep $P(Y|X)$ fixed; non-stationary AI systems change $P(Y|X)$ with the current input but have no internal memory; dynamic AI systems update $P(Y|X)$ through an internal memory state. On top of this classification it defines three properties: AIC robustness ($P(Y|X)$ is invariant over a pre-specified set of influencing factors), AIC sensitivity (changes in specified factors produce a KL-divergence response within a specified band), and AID stability (for dynamic systems, the KL divergence between successive output distributions never increases after a disturbance). The intended result is a generic safety-analysis basis that is architecture- and application-agnostic, refinable to specific systems, and usable from development through deployment and monitoring.

Load-bearing premise

The whole framework stands on the premise that the statistical input-output relationship $P(Y|X)$ is a sufficient and meaningful representation of an AI system's state for safety analysis; if internal state, memory, or dynamics can change safety-relevant behavior without changing $P(Y|X)$, the classifications and properties lose their foundation.

Editorial extensions

If this is right

Engineers can choose safety analyses by system class: static systems can be certified with fixed checks, non-stationary systems need context-aware checks, and dynamic systems require history-aware, temporal analysis.
The three properties give a specification language for safety that complements performance metrics and can be written into requirements, validation, and monitoring.
Online safety monitors can be built as input-output checks of AIC robustness, AIC sensitivity, and AID stability without needing internal state access.
The paradigm offers a common vocabulary that regulators and developers can refine per application, supporting lifecycle risk management as required by emerging AI regulation.
Dynamic AI systems with memory—such as recurrent and memory-augmented networks—can in principle be certified stable using only their output distribution sequence, which is testable at runtime.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The framework's reliance on $P(Y|X)$ as the state leaves open whether internal hidden states with no immediate effect on output still matter for future safety; a practical extension would condition stability criteria on a sufficient statistic of the memory state.
AID stability's KL-divergence condition is checkable in principle, but in high-dimensional output spaces KL estimates can be noisy; a testable extension would replace exact KL with a calibrated lower bound or a learned divergence proxy.
The static/non-stationary/dynamic trichotomy could be applied to today's foundation models: a fixed-context transformer is non-stationary, while a system with persistent memory is dynamic, which would change how their safety cases are structured.
The suggested enhancements—imagination of hazardous scenarios and moving-horizon retrospective confidence—could be combined into one architecture: an online reliability observer selects imagined risk cases for fallback planning.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear [name],

You should know two things about arXiv:2506.23703. It is a perspective paper, not a proof-based contribution. And its central formal property, AI Dynamics (AID) stability, does not do what it claims. The paper deserves a serious read because the conceptual frame is genuinely new, but the math needs a rework.

What is new: the authors propose a 'data control' paradigm in which AI systems are analyzed as data-based signals and systems, with three system classes (static, non-stationary, dynamic) and three properties (AIC robustness, AIC sensitivity, AID stability). They tie this to causal inference, dataset shift, and the EU AI Act, and they illustrate it on automated-driving tasks. The taxonomy is a useful lens; the tables are concrete and the writing is mostly clear.

Where it breaks down: Definition 3.3 defines AID stability as a monotonic decrease of D_KL[t] = D_KL(P_t || P_{t-1}), and then concludes that as t→∞ the system is stable. That inference is false. A non-negative decreasing sequence always converges to some limit, but the limit can be positive, and even if it is zero, consecutive distributions can approach each other while drifting without bound. A concrete example: take P_t = N(μ_t,1) with μ_t = √t. Then D_KL[t] ≈ 1/(8t), which is decreasing to zero, yet μ_t → ∞. So the condition (16) holds for all t, and the output distribution never settles. This is a counterexample to monotonic decay as a Lyapunov stability criterion. The paper also never shows that P(Y|X)_t has an equilibrium or that D_KL is a valid Lyapunov function with respect to that equilibrium. This is more than a technical slip; it undermines the core claim that AID stability provides a sound basis for safety analysis.

There are smaller issues: a duplicate Definition 2.2, a repeated paragraph in Section VI-B, and Definition 2.4 that mixes a discrete equality with a constant derivative (should be zero). The deeper assumption that P(Y|X) is sufficient to capture the state of a dynamic AI system is questionable, and the paper even acknowledges internal memory states h_t, yet the stability property looks only at input-output distributions.

Which is not to say the paper is worthless. As a perspective, it is a thoughtful synthesis, and the three classes and three properties are a sensible way to organize the discussion. If the authors reframe AID stability as a finite-horizon monitoring heuristic rather than a Lyapunov stability guarantee, the paper becomes much stronger. Who is this for: researchers who want a high-level map of control-theoretic concepts to AI safety, especially in automated driving. It is a discussion piece, not a technical reference.

Recommendation: if this came through my desk, I would send it to peer review, but expect major revision. The main referee point should be Definition 3.3. Fix that or re-scope it, and the rest becomes a solid perspective.

Referee Report

3 major / 5 minor

Summary. The paper proposes a new perspective on AI safety called "data control," built on an interdisciplinary fusion of data engineering and control theory. It introduces three AI system classes (static, non-stationary, dynamic), three system properties (AI Circumstance Robustness, AI Circumstance Sensitivity, AI Dynamics Stability), and several lifecycle-oriented enhancement mechanisms (enhanced assumption validation, enabled online system analysis, and responsible self-aware AI systems). The authors state that the main contribution is a generic, control-theoretic foundation for the safety analysis and assurance of AI systems, applicable across applications and architectures. The paper is qualitative and programmatic: it provides definitions and conceptual frameworks rather than proofs or experiments, and it explicitly defers evaluation to future work.

Significance. If the framework were made formally sound, the paper would represent a useful synthesis: it connects causality, dataset shift, statistical learning, and control-theoretic system analysis in one vocabulary, and it grounds the discussion in concrete automated-driving applications and EU AI Act requirements. The explicit definitions of system classes and properties are a step toward making AI safety analysis more systematic and less method-specific. The paper also ships an honest statement of limitations and future work, and its three proposed lifecycle mechanisms (especially the "imaginable space control" and the MHE/MPC-inspired reasoning) are original and potentially generative. However, the formal core is not yet reliable: the central stability definition is mathematically invalid, and the statistical dynamics principle is stated without a rigorous mathematical foundation. Given the paper's stated goal of providing a "generic basis for safety analysis and assurance," these flaws must be repaired before the framework can serve its purpose.

major comments (3)

[Definition 3.3 / Section VII-C, Eq. (16)] The AID stability definition is mathematically invalid. The condition D_KL[t+1] - D_KL[t] <= 0 only guarantees that the nonnegative sequence D_KL[t] converges to some L >= 0. If L > 0, the system continues to change indefinitely; if L = 0, the distributions can drift without limit. For example, let P(Y|X)_t be a unit-variance Gaussian with mean mu_t = sqrt(t). Then D_KL[t] = (mu_t - mu_{t-1})^2/2 approx 1/(8t), which is non-increasing and converges to zero, so Eq. (16) holds, while mu_t -> infinity and the output distribution does not converge to any limit. Consequently, the conclusion "for lim t -> infinity the dynamic AI system is stable" does not follow. The analogy to discrete-time Lyapunov stability (Eq. (17)) is also not valid because D_KL[t] is a distance between consecutive output distributions, not a positive-definite function of a well-defined state with a unique equilibrium. This is a load-bearing flaw: stability is one of the three central properties of the proposed data control paradigm, and the current definition does not provide a sound basis for safety analysis.
[Section V-C, Eq. (5)] The Principle of Statistical Dynamics is not mathematically well-posed. The expression dP(Y|X)/dt = F(X, CAU, LV, CF, I, CTX, TSK, E, t) treats a family of conditional probability distributions as a differentiable state, but no functional-analytic setting is provided: there is no specified metric or topology on the space of conditional distributions, no notion of differentiability, and no identification of a trajectory in that space. Since later definitions (Definitions 2.2-2.4 and 3.3) rely on P(Y|X)_t as a state vector, the formalization is fragile. The paper may legitimately propose this as a conceptual heuristic, but it should explicitly state that it is a perspective-level abstraction, not a formal equation, or it should supply the needed regularity assumptions.
[Definition 2.4 / Eq. (12)] The definition of a static AI system is internally inconsistent. Eq. (12) states P(Y|X)_{t+1} = P(Y|X)_t for all t, which implies the derivative dP(Y|X)/dt is zero. However, the immediately following text says "The underlying dynamic process is therefore assumed to be stationary: dP(Y|X)/dt = constant." An arbitrary constant is incompatible with the equality of Eq. (12); at most a zero constant would be consistent. This ambiguity affects the classification basis: readers cannot tell whether the intended condition is strict constancy of the conditional distribution or merely a time-homogeneous evolution. Please clarify the definition and align the prose with the equation.

minor comments (5)

[Throughout] Definition numbering is inconsistent: Definition 2.2 is used twice (for AI and for Dynamic AI System), and the later definitions are numbered 2.3 and 2.4 out of sequence. Please renumber all definitions sequentially.
[Section III-C] The paragraph beginning "Another data-based option to overcome the challenges is statistical learning [123]–[126]" appears to be accidentally duplicated with a slightly different reference [128] in the following sentence. Please remove the repetition and reconcile the citations.
[Eq. (10)] Equation (10) contains unmatched parentheses, e.g., in P(x_{t+1}|f_alpha(h_{t-1}, P(Y|X)_{t-1})) . Please check the derivation and fix the notation.
[Table VI] There is a typo in the table: "object detecor" should be "object detector."
[Section VII-C / Figure 7] Figure 7 is captioned as a visual illustration of AID stability, but the axes and the meaning of the omega symbols are not explained. Please add an explicit description of what is plotted, the role of omega, and how the plotted quantity relates to D_KL[t].

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the data-control framework is built from explicit postulates and definitions, with self-citations only as illustrative examples.

full rationale

The paper proposes a conceptual framework rather than deriving quantitative predictions from fitted data. Its central elements—CSKC, CAP, and the Principle of Statistical Dynamics—are introduced as explicit postulates or corollaries in Section V, not derived from the safety conclusions they later support. The AI system classes in Section VI and properties in Section VII (AIC robustness, AIC sensitivity, AID stability) are definitions that refine those postulates, so there is no estimation step that is later renamed as a prediction. No parameter is fitted to a subset of data and then reported as an independent result, and no external benchmark is claimed. The self-citations, e.g., [155] in Table VII, [257] in Table X, and [84]/[87] in the related work, serve only as illustrative applications or contextual pointers; they do not carry the load of the framework's central definitions. The AID stability condition (Eq. 16) is a proposed Lyapunov-style criterion, and whether non-increasing KL divergence genuinely implies convergence is a mathematical correctness concern, not a circularity concern, because the criterion is defined rather than derived from an input. Therefore, the derivation chain is self-contained at the conceptual level, and the minor self-citations are not load-bearing.

Assumptions & free parameters 0 free parameters · 4 assumptions · 2 invented entities

The paper does not fit parameters to data. Its formal content consists of new postulates and definitions. The main load-bearing axioms are the author-introduced CSKC, CAP, and PSD statements, plus the assumption that KL divergence over P(Y|X) can serve as a Lyapunov function. These are not derived from external benchmarks.

assumptions (4)

domain assumption CSKC: A learned causal model inherits the structural knowledge of the underlying ODE, so G(M_caus(X)) is a subset of Sigma(D).
Postulated in Eq. (2) without proof; used to link ODE models and causal models as sharing structural knowledge.
domain assumption CAP: The parameter set of a statistical model, mapped through S_stat, includes the mapped meaning of the ODE parameter set through S_ODE.
Postulated in Eq. (4); the mapping operators S_stat and S_ODE are not defined, making the claim untestable as stated.
domain assumption PSD: The conditional input-output distribution P(Y|X) evolves according to a dynamic process with causal, latent, confounder, intervention, context, task, and environment factors.
Stated as Eq. (5) without a derivation; it is the basis for the system class definitions.
domain assumption KL divergence over P(Y|X) is an admissible Lyapunov function for a dynamic AI system.
Assumed in Definition 3.3, Eqs. (16) and (17); no proof shows that the KL divergence between successive conditional distributions defines a Lyapunov function for the internal memory state.

invented entities (2)

Data control paradigm
purpose: Umbrella framework for AI safety analysis and assurance across the AI lifecycle.
The paper itself notes that realization and evaluation are future research; no falsifiable prediction is made outside the framework.
Imaginable space control (hazard case imaginator, probability evaluator, high-risk case selector)
purpose: Generates hypothetical hazard scenarios to feed to an AI when inputs are out of specification.
Presented as a conceptual architecture in Figure 8 with no implementation or test.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A New Perspective On AI Safety Through Control Theory Methodologies." pith.science (2026). https://pith.science/paper/KX3UX37C

@misc{pith2026250623703,
  author       = {Pith},
  title        = {Pith review of: A New Perspective On AI Safety Through Control Theory Methodologies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KX3UX37C}},
  note         = {Machine review of arXiv:2506.23703}
}

read the original abstract

While artificial intelligence (AI) is advancing rapidly and mastering increasingly complex problems with astonishing performance, the safety assurance of such systems is a major concern. Particularly in the context of safety-critical, real-world cyber-physical systems, AI promises to achieve a new level of autonomy but is hampered by a lack of safety assurance. While data-driven control takes up recent developments in AI to improve control systems, control theory in general could be leveraged to improve AI safety. Therefore, this article outlines a new perspective on AI safety based on an interdisciplinary interpretation of the underlying data-generation process and the respective abstraction by AI systems in a system theory-inspired and system analysis-driven manner. In this context, the new perspective, also referred to as data control, aims to stimulate AI engineering to take advantage of existing safety analysis and assurance in an interdisciplinary way to drive the paradigm of data control. Following a top-down approach, a generic foundation for safety analysis and assurance is outlined at an abstract level that can be refined for specific AI systems and applications and is prepared for future innovation.

Figures

Figures reproduced from arXiv: 2506.23703 by the authors.

**Figure 2.** The illustration presents the overall structure of the [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** AI system according to OCED [101]. the OECD. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: Comparison of causal and anti-causal AI systems [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Representation of a causal graph G based on the transformation of the ordinary differential equations into structural causal model according to [184]. In contrast, statistical models either neglect the structural knowledge as in classical statistical models like MLPs …

**Figure 6.** Figure 6: Illustration of AIC sensitivity. TABLE VII: Overview of AIC sensitivity across task-specific domains in automated driving. Task Method Description Scene Understanding Motion Prediction An appropriate context understanding requires the statistical input-output relati…

**Figure 7.** Figure 7: Visual illustration of AID stability. TABLE VIII: Overview of AID stability across task-specific domains in automated driving. Task Method Description Perception Object Tracking Dynamic multi-object tracking (MOT) AI systems, e.g., [228], aggregate target motion dyna…

**Figure 8.** Figure 8: Conceptualization of an extended input inspection including technical measures as risk management for foreseeable [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]

**Figure 9.** Figure 9: Concept of an online system analysis of observable AI [PITH_FULL_IMAGE:figures/full_fig_p019_9.png]

**Figure 10.** Figure 10: Visualization of a concept for the integration of ret [PITH_FULL_IMAGE:figures/full_fig_p020_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

273 extracted references · 59 canonical work pages

[1]

Object Classification Using CNN-Based Fusion of Vision and LIDAR in Autonomous Vehicle Environment,

H. Gao, B. Cheng, J. Wanget al., “Object Classification Using CNN-Based Fusion of Vision and LIDAR in Autonomous Vehicle Environment,”IEEE Trans. Ind. Inform., vol. 14, no. 9, pp. 4224–4231, 2018

2018
[2]

Toward Explainable Artificial Intelligence for Regression Models: A methodological perspective,

S. Letzgus, P. Wagner, J. Ledereret al., “Toward Explainable Artificial Intelligence for Regression Models: A methodological perspective,” IEEE Signal Process. Mag., vol. 39, no. 4, pp. 40–58, 2022

2022
[3]

A comprehensive survey of clustering algorithms: State-of-the-art machine learning ap- plications, taxonomy, challenges, and future research prospects,

A. E. Ezugwu, A. M. Ikotun, O. O. Oyeladeet al., “A comprehensive survey of clustering algorithms: State-of-the-art machine learning ap- plications, taxonomy, challenges, and future research prospects,”Eng. Appl. Artif. Intell., vol. 110, pp. 1–43, 2022, Art. no. 104743

2022
[4]

Time-series forecasting with deep learning: a survey,

B. Lim and S. Zohren, “Time-series forecasting with deep learning: a survey,”Phil. Trans. R. Soc., vol. 379, no. 2194, pp. 1–14, 2021, Art. no. 20200209

2021
[5]

Application of artificial intelligence algo- rithms in image processing,

X. Zhang and W. Dahu, “Application of artificial intelligence algo- rithms in image processing,”J. Vis. Commun. Image Represent., vol. 61, pp. 42–49, 2019

2019
[6]

An introduction to Deep Learning in Natural Language Processing: Models, techniques, and tools,

I. Lauriola, A. Lavelli, and F. Aiolli, “An introduction to Deep Learning in Natural Language Processing: Models, techniques, and tools,”Neurocomputing, vol. 470, pp. 443–456, 2022

2022
[7]

Causal inference and the data-fusion problem,

E. Bareinboim and J. Pearl, “Causal inference and the data-fusion problem,”Proc. Natl. Acad. Sci., vol. 113, no. 27, pp. 7345–7352, 2016

2016
[8]

A Comprehensive Survey on Transfer Learning,

F. Zhuang, Z. Qi, K. Duanet al., “A Comprehensive Survey on Transfer Learning,”Proc. IEEE, vol. 109, no. 1, pp. 43–76, 2020

2020

Show all 273 references

[9]

Supervised Learning,

P. Cunningham, M. Cord, and S. J. Delany, “Supervised Learning,” inMachine Learning Techniques for Multimedia: Case Studies on Organization and Retrieval. Springer, 2008, pp. 21–49

2008
[10]

Unsupervised Learning,

T. Hastie, R. Tibshirani, J. Friedmanet al., “Unsupervised Learning,” The Elements of Statistical Learning: Data Mining, Inference, and Prediction, pp. 485–585, 2009

2009
[11]

R. S. Sutton and A. G. Barto,Reinforcement Learning: An Introduction, 2nd ed. Cambridge, MA, US: MIT Press, 2018

2018
[12]

Jebara,Machine Learning: Discriminative and Generative

T. Jebara,Machine Learning: Discriminative and Generative. Springer Science & Business Media, 2012, vol. 755

2012
[13]

Generative ai for autonomous driving: Frontiers and opportunities,

Y . Wang, S. Xing, C. Canet al., “Generative ai for autonomous driving: Frontiers and opportunities,”arXiv preprint arXiv:2505.08854, 2025

2025 arXiv
[14]

Quo vadis artificial intelligence?

Y . Jiang, X. Li, H. Luoet al., “Quo vadis artificial intelligence?” Discover Artificial Intelligence, vol. 2, no. 4, pp. 1–19, 2022

2022
[15]

A. V . Oppenheim, A. S. Willsky, S. H. Nawabet al.,Signals and Systems. Prentice hall Upper Saddle River, NJ, 1997, vol. 2

1997
[16]

Girod, R

B. Girod, R. Rabenstein, and A. K. Stenger,Einf ¨uhrung in die Systemtheorie: Signale und Systeme in der Elektrotechnik und Infor- mationstechnik. Springer-Verlag, 2013

2013
[17]

W. J. Dally and J. W. Poulton,Digital Systems Engineering. Cam- bridge university press, 1998

1998
[18]

C. L. Phillips and H. T. Nagle,Digital Control System Analysis and Design. Prentice Hall Press, 2007

2007
[19]

The rise of artificial intelligence and the uncertain future for physicians,

C. Krittanawong, “The rise of artificial intelligence and the uncertain future for physicians,”Eur. J. Intern. Med., vol. 48, pp. e13–e14, 2018

2018
[20]

A comprehensive review on au- tomation in agriculture using artificial intelligence,

K. Jha, A. Doshi, P. Patelet al., “A comprehensive review on au- tomation in agriculture using artificial intelligence,”Artif. Intell. Agric., vol. 2, pp. 1–12, 2019

2019
[21]

The rise of artificial intelligence in healthcare applications,

A. Bohr and K. Memarzadeh, “The rise of artificial intelligence in healthcare applications,” inArtif. Intell. Healthcare. Elsevier, 2020, pp. 25–60

2020
[22]

Artificial Intelligence and Machine Learning Applications in Smart Production: Progress, Trends, and Directions,

R. Cioffi, M. Travaglioni, G. Piscitelliet al., “Artificial Intelligence and Machine Learning Applications in Smart Production: Progress, Trends, and Directions,”Sustainability, vol. 12, no. 2, p. 492, 2020

2020
[23]

Effectiveness of artificial intelligence techniques against cyber security risks apply of IT industry,

B. Alhayani, H. J. Mohammed, I. Z. Chaloobet al., “Effectiveness of artificial intelligence techniques against cyber security risks apply of IT industry,”Materials Today: Proceedings, vol. 531, 2021

2021
[24]

Artificial intelligence in sustainable energy industry: Status Quo, challenges and opportunities,

T. Ahmad, D. Zhang, C. Huanget al., “Artificial intelligence in sustainable energy industry: Status Quo, challenges and opportunities,” J. Clean. Prod., vol. 289, pp. 1–31, 2021, Art. no. 125834

2021
[25]

Developing artificial neural networks for safety critical systems,

Z. Kurd, T. Kelly, and J. Austin, “Developing artificial neural networks for safety critical systems,”Neural Comput. Appl., vol. 16, pp. 11–19, 2007

2007
[26]

Challenges in Using Neural Networks in Safety-Critical Applications,

H. Forsberg, J. Lind ´en, J. Hjorthet al., “Challenges in Using Neural Networks in Safety-Critical Applications,” inAIAA/IEEE Digit. Avion. Syst. Conf. - Proc. (DASC), 2020, pp. 1–7

2020
[27]

Systematic review of research on artificial intelligence applications in higher education– where are the educators?

O. Zawacki-Richter, V . I. Mar ´ın, M. Bondet al., “Systematic review of research on artificial intelligence applications in higher education– where are the educators?”Int. J. Educ. Technol. High. Educ., vol. 16, no. 1, pp. 1–27, 2019

2019
[28]

Financial time series forecasting with deep learning : A systematic literature review: 2005–2019,

O. B. Sezer, M. U. Gudelek, and A. M. Ozbayoglu, “Financial time series forecasting with deep learning : A systematic literature review: 2005–2019,”Appl. Soft Comput. J., vol. 90, pp. 1–32, 2020, Art. no. 106181

2005
[29]

Artificial Intelligence in Education: A Review,

L. Chen, P. Chen, and Z. Lin, “Artificial Intelligence in Education: A Review,”IEEE Access, vol. 8, pp. 75 264–75 278, 2020

2020
[30]

Artificial intelligence applications in solid waste management: A systematic research review,

M. Abdallah, M. A. Talib, S. Ferozet al., “Artificial intelligence applications in solid waste management: A systematic research review,” Waste Manag., vol. 109, pp. 231–246, 2020

2020
[31]

Developments, application, and performance of artificial intelligence in dentistry–A systematic review,

S. B. Khanagar, A. Al-Ehaideb, P. C. Maganuret al., “Developments, application, and performance of artificial intelligence in dentistry–A systematic review,”J. Dent. Sci., vol. 16, no. 1, pp. 508–522, 2021

2021
[32]

An embarrassingly simple approach to zero-shot learning,

B. Romera-Paredes and P. Torr, “An embarrassingly simple approach to zero-shot learning,” inProc. 32nd Int. Conf. Mach. Learn. (ICML). PMLR, 2015, pp. 2152–2161

2015
[33]

Matching Networks for One Shot Learning,

O. Vinyals, C. Blundell, T. Lillicrapet al., “Matching Networks for One Shot Learning,” inProc. 29th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2016

2016
[34]

Review and Analysis of Zero, One and Few Shot Learning Approaches,

S. Kadam and V . Vaidya, “Review and Analysis of Zero, One and Few Shot Learning Approaches,” inProc. 18th Int. Conf. Intell. Syst. Design Appl. (ISDA). Springer, 2018, pp. 100–112

2018
[35]

Meta-learning,

J. Vanschoren, “Meta-learning,” inAutomated Machine Learning: Methods, Systems, Challenges. Springer, 2019, pp. 35–61

2019
[36]

Safety Assurance of Artificial Intelligence-Based Systems: A Systematic Literature Review on the State of the Art and Guidelines for Future Work,

A. V . S. Neto, J. B. Camargo, J. R. Almeidaet al., “Safety Assurance of Artificial Intelligence-Based Systems: A Systematic Literature Review on the State of the Art and Guidelines for Future Work,”IEEE Access, vol. 10, pp. 130 733–130 770, 2022

2022
[37]

Guaranteed Safe Online Learning via Reachability: tracking a ground target using a quadrotor,

J. H. Gillula and C. J. Tomlin, “Guaranteed Safe Online Learning via Reachability: tracking a ground target using a quadrotor,” inProc. IEEE Int. Conf. Robot. Autom. (ICRA), 2012, pp. 2723–2730

2012
[38]

A Survey of Algorithms for Black-Box Safety Validation of Cyber-Physical Systems,

A. Corso, R. Moss, M. Korenet al., “A Survey of Algorithms for Black-Box Safety Validation of Cyber-Physical Systems,”J. Artif. Intell. Res., vol. 72, pp. 377–428, 2021

2021
[39]

DeepXplore: Automated Whitebox Testing of Deep Learning Systems,

K. Pei, Y . Cao, J. Yanget al., “DeepXplore: Automated Whitebox Testing of Deep Learning Systems,” inProc. 26th ACM Symp. Oper. Syst. Princ., 2017, pp. 1–18

2017
[40]

TensorFI: A Flexible Fault Injection Framework for TensorFlow Applications,

Z. Chen, N. Narayanan, B. Fanget al., “TensorFI: A Flexible Fault Injection Framework for TensorFlow Applications,” inProc. IEEE 31st Int. Symp. on Software Reliability Engineering (ISSRE), 2020, pp. 426– 435

2020
[41]

Jespipe: A Plugin-Based, Open MPI Framework for Adversarial Machine Learning Analysis,

S. Alemany, J. Nucciarone, and N. Pissinou, “Jespipe: A Plugin-Based, Open MPI Framework for Adversarial Machine Learning Analysis,” in Proc. IEEE Int. Conf. Big Data (Big Data). IEEE, 2021, pp. 3663– 3670

2021
[42]

G. Trusted-AI, “GitHub Trusted-AI/Adversarial-Robustness-Toolbox: Adversarial Robustness Toolbox (ART) Python Library for Machine Learning Security Evasion, Poisoning, Extraction, Inference Red and Blue Teams,” 2022. [Online]. Available: https://github.com/TrustedAI /adversari...

2022
[43]

SMOF: A safety monitoring framework for autonomous systems,

M. Machin, J. Guiochet, H. Waeselyncket al., “SMOF: A safety monitoring framework for autonomous systems,”IEEE Trans. Syst. Man Cybern.: Syst., vol. 48, no. 5, pp. 702–715, 2016

2016
[44]

Considerations of Artificial Intelligence Safety Engineering for Unmanned Aircraft,

S. Schirmer, C. Torens, F. Nikodemet al., “Considerations of Artificial Intelligence Safety Engineering for Unmanned Aircraft,” inProc. 37th Int. Conf. on Computer Safety, Reliability, and Security (SAFECOMP). Springer, 2018, pp. 465–472

2018
[45]

The Role of Explainability in Assuring Safety of Machine Learning in Healthcare,

Y . Jia, J. McDermid, T. Lawtonet al., “The Role of Explainability in Assuring Safety of Machine Learning in Healthcare,”IEEE Trans. Emerg. Top. Comput., vol. 10, no. 4, pp. 1746–1760, 2022

2022
[46]

DLOAM: Real-time and Robust LiDAR SLAM System Based on CNN in Dynamic Urban Environments,

W. Liu, W. Sun, and Y . Liu, “DLOAM: Real-time and Robust LiDAR SLAM System Based on CNN in Dynamic Urban Environments,”IEEE Open J. Intell. Transp. Syst., 2021

2021
[47]

G. C. Goodwin, S. F. Graebe, M. E. Salgadoet al.,Control System Design. NJ, USA: Prentice-Hall, 2001

2001
[48]

Lunze,Regelungstechnik 1: Systemtheoretische Grundlagen, Analyse und Entwurf einschleifiger Regelungen

J. Lunze,Regelungstechnik 1: Systemtheoretische Grundlagen, Analyse und Entwurf einschleifiger Regelungen. Springer, 2016, vol. 10

2016
[49]

Laplace-, Fourier-und z-Transformation, 10., ¨uberarb,

O. F ¨ollinger and K. Mathias, “Laplace-, Fourier-und z-Transformation, 10., ¨uberarb,”10. Aufl. VDE-Verlag, Berlin, 2011

2011
[50]

M. S. Fadali and A. Visioli,Digital Control Engineering: Analysis and Design. Amsterdam, Netherlands: Elsevier Science, 2012

2012
[51]

N. S. Nise,Control Systems Engineering. John Wiley & Sons, 2020

2020
[52]

An overview of multi-task learning,

Y . Zhang and Q. Yang, “An overview of multi-task learning,”National Science Review, vol. 5, no. 1, pp. 30–43, 2018

2018
[53]

Neural Lyapunov Control,

Y .-C. Chang, N. Roohi, and S. Gao, “Neural Lyapunov Control,” in Proc. 32nd Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2019, pp. 1–10

2019
[54]

Data-Driven Model Predictive Control With Stability and Robustness Guarantees,

J. Berberich, J. K ¨ohler, M. A. M ¨ulleret al., “Data-Driven Model Predictive Control With Stability and Robustness Guarantees,”IEEE Trans. Autom. Control, vol. 66, no. 4, pp. 1702–1717, 2020

2020
[55]

Training Robust Neural Networks Using Lipschitz Bounds,

P. Pauli, A. Koch, J. Berberichet al., “Training Robust Neural Networks Using Lipschitz Bounds,”IEEE Control Syst. Lett., vol. 6, pp. 121–126, 2021

2021
[56]

A predictive safety filter for learning-based control of constrained nonlinear dynamical systems,

K. P. Wabersich and M. N. Zeilinger, “A predictive safety filter for learning-based control of constrained nonlinear dynamical systems,” Automatica, vol. 129, p. 109597, 2021

2021
[57]

Deep Online Learning Via Meta-Learning: Continual Adaptation for Model-Based RL,

A. Nagabandi, C. Finn, and S. Levine, “Deep Online Learning Via Meta-Learning: Continual Adaptation for Model-Based RL,” inInt. Conf. Learn. Represent. (ICLR), 2018, pp. 1–15

2018
[58]

Online Reinforcement Learning in Stochastic Games,

C.-Y . Wei, Y .-T. Hong, and C.-J. Lu, “Online Reinforcement Learning in Stochastic Games,” inProc. 30th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2017, pp. 1–11

2017
[59]

Online reinforcement learning for a continuous space system with experimental validation,

O. Dogru, N. Wieczorek, K. Velswamyet al., “Online reinforcement learning for a continuous space system with experimental validation,” J. Process Control, vol. 104, pp. 86–100, 2021

2021
[60]

Safe Model-based Reinforcement Learning with Stability Guarantees,

F. Berkenkamp, M. Turchetta, A. Schoelliget al., “Safe Model-based Reinforcement Learning with Stability Guarantees,” inProc. 31st Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2017, pp. 1–11

2017
[61]

Towards Safe Online Rein- forcement Learning in Computer Systems,

H. Mao, M. Schwarzkopf, H. Heet al., “Towards Safe Online Rein- forcement Learning in Computer Systems,” inProc. 33rd Int. Conf. Neural Inf. Process. Syst. (NeurIPS), Machine Learning for Systems Workshop, 2019, pp. 1–9

2019
[62]

Robustness and Adaptability of Reinforcement Learning-Based Cooperative Autonomous Driving in Mixed-Autonomy Traffic,

R. Valiente, B. Toghi, R. Pedarsaniet al., “Robustness and Adaptability of Reinforcement Learning-Based Cooperative Autonomous Driving in Mixed-Autonomy Traffic,”IEEE Open J. Intell. Transp. Syst., vol. 3, pp. 397–410, 2022

2022
[63]

Learning-Based Safe Control for Robot and Autonomous Vehicle Using Efficient Safety Certificate,

H. Zheng, C. Chen, S. Liet al., “Learning-Based Safe Control for Robot and Autonomous Vehicle Using Efficient Safety Certificate,” IEEE Open J. Intell. Transp. Syst., vol. 4, pp. 419–430, 2023

2023
[64]

Data-Driven Safety Filters: Hamilton-Jacobi Reachability, Control Barrier Functions, and Predictive Methods for Uncertain Systems,

K. P. Wabersich, A. J. Taylor, J. J. Choiet al., “Data-Driven Safety Filters: Hamilton-Jacobi Reachability, Control Barrier Functions, and Predictive Methods for Uncertain Systems,”IEEE Control Syst. Mag., vol. 43, no. 5, pp. 137–177, 2023

2023
[65]

Predictive safety filter using system level synthesis,

A. Leeman, J. K ¨ohler, S. Bennaniet al., “Predictive safety filter using system level synthesis,” inProc. 5th Annual Learning for Dynamics and Control Conference (L4DC). PMLR, 2023, pp. 1180–1192

2023
[66]

Model Predictive Control,

B. Kouvaritakis and M. Cannon, “Model Predictive Control,”Switzer- land: Springer International Publishing, vol. 38, pp. 13–56, 2016

2016
[67]

From model-based control to data-driven control: Survey, classification and perspective,

Z.-S. Hou and Z. Wang, “From model-based control to data-driven control: Survey, classification and perspective,”Inf. Sci., vol. 235, pp. 3–35, 2013

2013
[68]

An Overview of Dynamic-Linearization- Based Data-Driven Control and Applications,

Z. Hou, R. Chi, and H. Gao, “An Overview of Dynamic-Linearization- Based Data-Driven Control and Applications,”IEEE Trans. Ind. Elec- tron., vol. 64, no. 5, pp. 4076–4090, 2016

2016
[69]

Data-driven control: A behavioral approach,

T. M. Maupong and P. Rapisarda, “Data-driven control: A behavioral approach,”Syst. Control Lett., vol. 101, pp. 37–43, 2017

2017
[70]

Formulas for Data-Driven Control: Stabiliza- tion, Optimality, and Robustness,

C. De Persis and P. Tesi, “Formulas for Data-Driven Control: Stabiliza- tion, Optimality, and Robustness,”IEEE Trans. Autom. Control, vol. 65, no. 3, pp. 909–924, 2019

2019
[71]

Data-driven mpc for quadrotors,

G. Torrente, E. Kaufmann, P. F ¨ohnet al., “Data-driven mpc for quadrotors,”IEEE Robot. Autom. Lett., vol. 6, no. 2, pp. 3769–3776, 2021

2021
[72]

Behavioral systems theory in data-driven analysis, signal processing, and control,

I. Markovsky and F. D ¨orfler, “Behavioral systems theory in data-driven analysis, signal processing, and control,”Annual Reviews in Control, vol. 52, pp. 42–64, 2021

2021
[73]

Learning model predictive control for iterative tasks. a data-driven control framework,

U. Rosolia and F. Borrelli, “Learning model predictive control for iterative tasks. a data-driven control framework,”IEEE Trans. Autom. Control, vol. 63, no. 7, pp. 1883–1896, 2017

2017
[74]

Bridging Direct and Indirect Data-Driven Control Formulations via Regularizations and Relaxations,

F. D ¨orfler, J. Coulson, and I. Markovsky, “Bridging Direct and Indirect Data-Driven Control Formulations via Regularizations and Relaxations,”IEEE Trans. Autom. Control, vol. 68, no. 2, pp. 883– 897, 2022

2022
[75]

Adaptive Observer Based Data-Driven Control for Nonlinear Discrete-Time Processes,

D. Xu, B. Jiang, and P. Shi, “Adaptive Observer Based Data-Driven Control for Nonlinear Discrete-Time Processes,”IEEE Trans. Autom. Sci. Eng., vol. 11, no. 4, pp. 1037–1045, 2013

2013
[76]

A Behavioral Approach to Data-Driven Control With Noisy Input–Output Data,

H. J. van Waarde, J. Eising, M. K. Camlibelet al., “A Behavioral Approach to Data-Driven Control With Noisy Input–Output Data,” IEEE Trans. Autom. Control, 2023

2023
[77]

Flight Test Validation of a Safety-Critical Neural Network Based Longitudinal Controller for a Fixed-Wing UAS,

D. Shukla, R. Lal, D. Hauptmanet al., “Flight Test Validation of a Safety-Critical Neural Network Based Longitudinal Controller for a Fixed-Wing UAS,” inAIAA Aviation Forum, 2020, p. 3093

2020
[78]

On Causal and Anticausal Learning,

B. Sch ¨olkopf, D. Janzing, J. Peterset al., “On Causal and Anticausal Learning,”arXiv preprint arXiv:1206.6471, 2012

2012 arXiv
[79]

Covariate Shift Adap- tation by Importance Weighted Cross Validation,

M. Sugiyama, M. Krauledat, and K.-R. M ¨uller, “Covariate Shift Adap- tation by Importance Weighted Cross Validation,”J. Mach. Learn. Res., vol. 8, no. 5, 2007

2007
[80]

Domain Adaptation under Target and Conditional Shift,

K. Zhang, B. Sch ¨olkopf, K. Muandetet al., “Domain Adaptation under Target and Conditional Shift,” inProc. 32th Int. Conf. Mach. Learn. (ICML). PMLR, 2013, pp. 819–827

2013
[81]

Reliable Decision Support using Counter- factual Models,

P. Schulam and S. Saria, “Reliable Decision Support using Counter- factual Models,” inProc. 30th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2017, pp. 1–12

2017
[82]

Pearl,Causality

J. Pearl,Causality. Cambridge University Press, 2009

2009
[83]

Towards a multi-agent vision-language system for zero-shot novel hazardous object detection for autonomous driving safety,

S. Shriram, S. Perisetla, A. Keskaret al., “Towards a multi-agent vision-language system for zero-shot novel hazardous object detection for autonomous driving safety,”IEEE RAS Conference on Automation Science and Engineering, 2025

2025
[84]

Perception without vision for trajectory prediction: Ego vehicle dynamics as scene representation for efficient active learning in autonomous driving,

R. Greer and M. Trivedi, “Perception without vision for trajectory prediction: Ego vehicle dynamics as scene representation for efficient active learning in autonomous driving,”IEEE ITSS Intelligent Trans- portation Systems Conference, 2025

2025
[85]

Generalizing from a Few Examples: A Survey on Few-shot Learning,

Y . Wang, Q. Yao, J. T. Kwoket al., “Generalizing from a Few Examples: A Survey on Few-shot Learning,”ACM Comput. Surv., vol. 53, no. 3, pp. 1–34, 2020

2020
[86]

Language-driven active learning for diverse open-set 3d object detection,

R. Greer, B. Antoniussen, A. Møgelmoseet al., “Language-driven active learning for diverse open-set 3d object detection,” inProceedings of the Winter Conference on Applications of Computer Vision, 2025, pp. 980–988

2025
[87]

Towards explainable, safe autonomous driving with language embeddings for novelty identification and active learning: Framework and experimental analysis with real-world data sets,

R. Greer and M. Trivedi, “Towards explainable, safe autonomous driving with language embeddings for novelty identification and active learning: Framework and experimental analysis with real-world data sets,”arXiv preprint arXiv:2402.07320, 2024

2024 arXiv
[88]

Learning to Learn Using Gradient Descent,

S. Hochreiter, A. S. Younger, and P. R. Conwell, “Learning to Learn Using Gradient Descent,” inProc. 11st Int. Conf. Art. Neural Netw. (ICANN). Springer, 2001, pp. 87–94

2001
[89]

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks,

C. Finn, P. Abbeel, and S. Levine, “Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks,” inProc. 34th Int. Conf. Mach. Learn. (ICML). PMLR, 2017, pp. 1126–1135

2017
[90]

Data-driven distributionally ro- bust optimization using the Wasserstein metric: performance guarantees and tractable reformulations,

P. Mohajerin Esfahani and D. Kuhn, “Data-driven distributionally ro- bust optimization using the Wasserstein metric: performance guarantees and tractable reformulations,”Math. Program., vol. 171, no. 1, pp. 115– 166, 2018

2018
[91]

Certifiable Distributional Robustness with Principled Adversarial Training,

A. Sinha, H. Namkoong, R. V olpiet al., “Certifiable Distributional Robustness with Principled Adversarial Training,”arXiv preprint arXiv:1710.10571, 2017

2017 arXiv
[92]

Distributionally Robust Optimization: A Review,

H. Rahimian and S. Mehrotra, “Distributionally Robust Optimization: A Review,”arXiv preprint arXiv:1908.05659, 2019

1908 arXiv
[93]

From development to deployment: dataset shift, causality, and shift-stable models in health AI,

A. Subbaswamy and S. Saria, “From development to deployment: dataset shift, causality, and shift-stable models in health AI,”Biostatis- tics, vol. 21, no. 2, pp. 345–352, 2020

2020
[94]

Evaluating model robustness and stability to dataset shift,

A. Subbaswamy, R. Adams, and S. Saria, “Evaluating model robustness and stability to dataset shift,” inInternational conference on artificial intelligence and statistics. PMLR, 2021, pp. 2611–2619

2021
[95]

Concrete problems in ai safety,

D. Amodei, C. Olah, J. Steinhardtet al., “Concrete problems in ai safety,”arXiv preprint arXiv:1606.06565, 2016

2016 arXiv
[96]

Evaluation of CNN-Based Ap- proaches to Adverse Weather Image Classification for Autonomous Driving Systems,

V . Afxentiou and T. Vladimirova, “Evaluation of CNN-Based Ap- proaches to Adverse Weather Image Classification for Autonomous Driving Systems,”IEEE Open J. Intell. Transp. Syst., 2025

2025
[97]

Causal Modeling for Training and Evaluating Dataset Shift-Stable Machine Learning Models in Healthcare,

A. Subbaswamyet al., “Causal Modeling for Training and Evaluating Dataset Shift-Stable Machine Learning Models in Healthcare,” Ph.D. dissertation, Johns Hopkins University, 2023

2023
[98]

Learning repre- sentations by back-propagating errors,

D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning repre- sentations by back-propagating errors,”Nature, vol. 323, no. 6088, pp. 533–536, 1986

1986
[99]

Long Short-Term Memory,

S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,”Neural Comput., vol. 9, no. 8, pp. 1735–1780, 1997

1997
[100]

A Path Towards Autonomous Machine Intelligence Version 0.9. 2, 2022-06-27,

Y . LeCun, “A Path Towards Autonomous Machine Intelligence Version 0.9. 2, 2022-06-27,”Open Review, vol. 62, 2022

2022
[101]

Scoping the OECD AI principles,

OECD, “Scoping the OECD AI principles,”OECDpublishing, no. 291, 2019. [Online]. Available: https://www.oecd-ilibrary.org/content/ paper/d62f618a-en

2019
[102]

Robustness, security and safety (Principle 1.4),

——, “Robustness, security and safety (Principle 1.4),” 2025. [Online]. Available: https://oecd.ai/en/dashboards/ai-principles/P8

2025
[103]

International Scientific Report on the Safety of Advanced AI (Interim Report),

Y . Bengio, S. Mindermann, D. Priviteraet al., “International Scientific Report on the Safety of Advanced AI (Interim Report),”arXiv preprint arXiv:2412.05282, 2024

2024 arXiv
[104]

Safe ai–how is this possible?

H. Rueß and S. Burton, “Safe ai–how is this possible?”arXiv preprint arXiv:2201.10436, 2022

2022 arXiv
[105]

The United States Artificial Intelligence Safety Institute: Vision, Mission, and Strategic Goals,

N. I. of Standards and T. (NIST), “The United States Artificial Intelligence Safety Institute: Vision, Mission, and Strategic Goals,”
[106]

Responsible Artificial Intelligence Systems: A Roadmap to Society’s Trust through Trustworthy AI, Auditability, Accountability, and Governance,

A. Herrera-Poyatos, J. Del Ser, M. L. de Pradoet al., “Responsible Artificial Intelligence Systems: A Roadmap to Society’s Trust through Trustworthy AI, Auditability, Accountability, and Governance,”arXiv preprint arXiv:2503.04739, 2025

2025
[107]

Hendrycks,Introduction to AI Safety, Ethics, and Society

D. Hendrycks,Introduction to AI Safety, Ethics, and Society. Taylor & Francis, 2025

2025
[108]

Probabilities of Causation: Role of Observational Data,

A. Li and J. Pearl, “Probabilities of Causation: Role of Observational Data,” inProc. 26th Int. Conf. Artif. Intell. Statist. (AISTATS). PMLR, 2023, pp. 10 012–10 027

2023
[109]

Even small correlation and diversity shifts pose dataset-bias issues,

A. Bissoto, C. Barata, E. Valleet al., “Even small correlation and diversity shifts pose dataset-bias issues,”Pattern Recognit. Lett., 2024

2024
[110]

The Next Frontier: AI We Can Really Trust,

A. Holzinger, “The Next Frontier: AI We Can Really Trust,” inMachine Learning and Principles and Practice of Knowledge Discovery in Databases. Cham: Springer International Publishing, 2021, pp. 427– 440

2021
[111]

When Training and Test Sets Are Different: Charac- terizing Learning Transfer,

A. Storkey, “When Training and Test Sets Are Different: Charac- terizing Learning Transfer,” inDataset Shift in Machine Learning, J. Qui ˜nonero-Candela, M. Sugiyama, A. Schwaighoferet al., Eds. Cambridge, MA, USA: MIT Press, 2009, pp. 3–28

2009
[112]

Reichenbach,The Direction of Time

H. Reichenbach,The Direction of Time. Univ. of California Press, 1991, vol. 65

1991
[113]

Toward Causal Represen- tation Learning,

B. Sch ¨olkopf, F. Locatello, S. Baueret al., “Toward Causal Represen- tation Learning,”Proc. IEEE, vol. 109, no. 5, pp. 612–634, 2021

2021
[114]

A Review of Off-Line Mode Dataset Shifts,

C. C. Takahashi and A. P. Braga, “A Review of Off-Line Mode Dataset Shifts,”IEEE Comput. Intell. Mag., vol. 15, no. 3, pp. 16–27, 2020

2020
[115]

Domain Generalization: A Survey,

K. Zhou, Z. Liu, Y . Qiaoet al., “Domain Generalization: A Survey,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 4, pp. 4396–4415, 2022

2022
[116]

Improving predictive inference under covariate shift by weighting the log-likelihood function,

H. Shimodaira, “Improving predictive inference under covariate shift by weighting the log-likelihood function,”J. Stat. Plan. Inference, vol. 90, no. 2, pp. 227–244, 2000

2000
[117]

Direct Importance Estimation with Model Selection and Its Application to Covariate Shift Adaptation,

M. Sugiyama, S. Nakajima, H. Kashimaet al., “Direct Importance Estimation with Model Selection and Its Application to Covariate Shift Adaptation,” inProc. 20th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2007, pp. 1–8

2007
[118]

Dataset Shift Detection in Non- stationary Environments Using EWMA Charts,

H. Raza, G. Prasad, and Y . Li, “Dataset Shift Detection in Non- stationary Environments Using EWMA Charts,” inProc. IEEE Int. Conf. Syst. Man Cybern.: Syst., 2013, pp. 3151–3156

2013
[119]

Probabilistic Causality and Simpson’s Paradox,

R. Otte, “Probabilistic Causality and Simpson’s Paradox,”Philosophy of Science, vol. 52, no. 1, pp. 110–125, 1985

1985
[120]

A Theory of Causal Learning in Children: Causal Maps and Bayes Nets

A. Gopnik, C. Glymour, D. M. Sobelet al., “A Theory of Causal Learning in Children: Causal Maps and Bayes Nets.”Psychol. Rev., vol. 111, no. 1, p. 3, 2004

2004
[121]

Gopnik, L

A. Gopnik, L. Schulz, and L. E. Schulz,Causal Learning: Psychology, Philosophy, and Computation. Oxford University Press, 2007

2007
[122]

Causal Learning and Inference as a Rational Process: The New Synthesis,

K. J. Holyoak and P. W. Cheng, “Causal Learning and Inference as a Rational Process: The New Synthesis,”Annu. Rev. Psychol., vol. 62, pp. 135–163, 2011

2011
[123]

Challenging Common Assump- tions in the Unsupervised Learning of Disentangled Representations,

F. Locatello, S. Bauer, M. Lucicet al., “Challenging Common Assump- tions in the Unsupervised Learning of Disentangled Representations,” inProc. 36th Int. Conf. Mach. Learn. (ICML). PMLR, 2019, pp. 4114–4124

2019
[124]

Disentangled Representation Learning GAN for Pose-Invariant Face Recognition,

L. Tran, X. Yin, and X. Liu, “Disentangled Representation Learning GAN for Pose-Invariant Face Recognition,” inProc. IEEE/CVF Com- put. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2017, pp. 1415– 1424

2017
[125]

Disentangled Feature Representation for Few-Shot Image Classification,

H. Cheng, Y . Wang, H. Liet al., “Disentangled Feature Representation for Few-Shot Image Classification,”IEEE Trans. Neural Netw. Learn. Syst., 2023

2023
[126]

beta-V AE: Learning Basic Visual Concepts with a Constrained Variational Framework,

I. Higgins, L. Matthey, A. Palet al., “beta-V AE: Learning Basic Visual Concepts with a Constrained Variational Framework,” inInt. Conf. Learn. Represent. (ICLR), 2017, pp. 1–22

2017
[127]

Disentangling by Factorising,

H. Kim and A. Mnih, “Disentangling by Factorising,” inProc. 35th Int. Conf. Mach. Learn. (ICML). PMLR, 2018, pp. 2649–2658

2018
[128]

V . N. Vapnik, V . Vapniket al.,Statistical Learning Theory. New York, NY , USA: Wiley, 1998

1998
[129]

Multitask Learning,

R. Caruana, “Multitask Learning,”Machine Learning, vol. 28, pp. 41– 75, 1997

1997
[130]

A Survey of Zero-Shot Learning: Settings, Methods, and Applications,

W. Wang, V . W. Zheng, H. Yuet al., “A Survey of Zero-Shot Learning: Settings, Methods, and Applications,”ACM Trans. Intell. Syst. Technol., vol. 10, no. 2, 2019

2019
[131]

Unsupervised Domain Adaptation by Backpropagation,

Y . Ganin and V . Lempitsky, “Unsupervised Domain Adaptation by Backpropagation,” inProc. 32nd Int. Conf. Mach. Learn. (ICML). PMLR, 2015, pp. 1180–1189

2015
[132]

Maximum Classifier Dis- crepancy for Unsupervised Domain Adaptation,

K. Saito, K. Watanabe, Y . Ushikuet al., “Maximum Classifier Dis- crepancy for Unsupervised Domain Adaptation,” inProc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2018, pp. 3723–3732

2018
[133]

Adapting Visual Category Models to New Domains,

K. Saenko, B. Kulis, M. Fritzet al., “Adapting Visual Category Models to New Domains,” inProc. 11th IEEE Europ. Conf. Comp. Vision (ECCV). Springer, 2010, pp. 213–226

2010
[134]

A Survey on Transfer Learning,

S. J. Pan and Q. Yang, “A Survey on Transfer Learning,”IEEE Trans. Knowl. Data Eng., vol. 22, no. 10, pp. 1345–1359, 2009

2009
[135]

Generalizing from Several Related Classification Tasks to a New Unlabeled Sample,

G. Blanchard, G. Lee, and C. Scott, “Generalizing from Several Related Classification Tasks to a New Unlabeled Sample,” inProc. 24th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2011, pp. 1–9

2011
[136]

The Functional Basis Of Counterfactual Thinking,

N. J. Roese, “The Functional Basis Of Counterfactual Thinking,”J. Pers. Soc. Psychol., vol. 66, no. 5, p. 805, 1994

1994
[137]

Sample-Efficient Reinforcement Learning via Counterfactual-Based Data Augmentation,

C. Lu, B. Huang, K. Wanget al., “Sample-Efficient Reinforcement Learning via Counterfactual-Based Data Augmentation,”arXiv preprint arXiv:2012.09092, 2020

2012 arXiv
[138]

Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Search,

L. Buesing, T. Weber, Y . Zwolset al., “Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Search,”arXiv preprint arXiv:1811.06272, 2018

2018 arXiv
[139]

Activation Functions in Neural Networks,

S. Sharma, S. Sharma, and A. Athaiya, “Activation Functions in Neural Networks,”Int. J. Eng. Sci. Technol., vol. 6, no. 12, pp. 310–316, 2017

2017
[140]

A New Varying-Parameter Convergent-Differential Neural-Network for Solving Time-Varying Convex QP Problem Constrained by Linear-Equality,

Z. Zhang, Y . Lu, L. Zhenget al., “A New Varying-Parameter Convergent-Differential Neural-Network for Solving Time-Varying Convex QP Problem Constrained by Linear-Equality,”IEEE Trans. Autom. Control, vol. 63, no. 12, pp. 4110–4125, 2018

2018
[141]

A novel bearing intelligent fault diagnosis framework under time-varying working conditions using recurrent neural network,

Z. An, S. Li, J. Wanget al., “A novel bearing intelligent fault diagnosis framework under time-varying working conditions using recurrent neural network,”ISA Transactions, vol. 100, pp. 155–170, 2020

2020
[142]

Dynamic Neural Network Models for Time-Varying Problem Solving: A Survey on Model Structuress,

C. Hua, X. Cao, Q. Xuet al., “Dynamic Neural Network Models for Time-Varying Problem Solving: A Survey on Model Structuress,”IEEE Access, 2023

2023
[143]

An Introduction to Computational Geom- etry,

M. Minsky and S. Papert, “An Introduction to Computational Geom- etry,”Cambridge tiass., HIT, vol. 479, no. 480, p. 104, 1969

1969
[144]

Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position,

K. Fukushima, “Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position,” Biol. Cybern., vol. 36, no. 4, pp. 193–202, 1980

1980
[145]

Gradient-based learning applied to document recognition,

Y . LeCun, L. Bottou, Y . Bengioet al., “Gradient-based learning applied to document recognition,”Proc. IEEE, vol. 86, no. 11, pp. 2278–2324, 1998

1998
[146]

The Graph Neural Network Model,

F. Scarselli, M. Gori, A. C. Tsoiet al., “The Graph Neural Network Model,”IEEE Trans. Neural Netw. Learn. Syst., vol. 20, no. 1, pp. 61–80, 2008

2008
[147]

Autoencoders, Unsupervised Learning, and Deep Archi- tectures,

P. Baldi, “Autoencoders, Unsupervised Learning, and Deep Archi- tectures,” inProc. 35th Int. Conf. Mach. Learn. (ICML), Workshop on Unsupervised and Transfer Learning. JMLR Workshop and Conference Proceedings, 2012, pp. 37–49

2012
[148]

Auto-encoding variational bayes,

D. P. Kingma and M. Welling, “Auto-encoding variational bayes,” in Proc. 2nd Int. Conf. Learn. Represent. (ICLR), 2013

2013
[149]

Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication,

H. Jaeger and H. Haas, “Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication,”Science, vol. 304, no. 5667, pp. 78–80, 2004

2004
[150]

Learning phrase representations using RNN encoder-decoder for statistical machine translation,

K. Cho, B. Van Merri ¨enboer, C. Gulcehreet al., “Learning phrase representations using RNN encoder-decoder for statistical machine translation,”arXiv preprint arXiv:1406.1078, 2014

2014 arXiv
[151]

Echo state network,

H. Jaeger, “Echo state network,”scholarpedia, vol. 2, no. 9, p. 2330, 2007

2007
[152]

An online self-adaptive modular neural network for time-varying systems,

J. Qiao, Z. Zhang, and Y . Bo, “An online self-adaptive modular neural network for time-varying systems,”Neurocomputing, vol. 125, pp. 7– 16, 2014

2014
[153]

Adaptive Neural Network Learning Controller Design for a Class of Nonlinear Systems With Time-Varying State Constraints,

Y .-J. Liu, L. Ma, L. Liuet al., “Adaptive Neural Network Learning Controller Design for a Class of Nonlinear Systems With Time-Varying State Constraints,”IEEE Trans. Neural Netw. Learn. Syst., vol. 31, no. 1, pp. 66–75, 2019

2019
[154]

Novel Recurrent Neural Network for Time- Varying Problems Solving [Research Frontier],

D. Guo and Y . Zhang, “Novel Recurrent Neural Network for Time- Varying Problems Solving [Research Frontier],”IEEE Comput. Intell. Mag., vol. 7, no. 4, pp. 61–65, 2012

2012
[155]

Robust Meta-Learning of Vehicle Yaw Rate Dynamics via Conditional Neural Processes,

L. Ullrich, A. V ¨olz, and K. Graichen, “Robust Meta-Learning of Vehicle Yaw Rate Dynamics via Conditional Neural Processes,” in Proc. 62nd IEEE Conf. Decis. Control (CDC), 2023, pp. 2611–2619

2023
[156]

F ¨ollinger and D

O. F ¨ollinger and D. Franke,Einf ¨uhrung in die Zustandsbeschreibung dynamischer Systeme. M ¨unchen: Oldenbourg, 1982

1982
[157]

LMI-based approach for asymptotically stability analysis of delayed neural networks,

X. Liao, G. Chen, and E. N. Sanchez, “LMI-based approach for asymptotically stability analysis of delayed neural networks,”IEEE Trans. Circuits Syst. I: Fundamental Theory and Applications, vol. 49, no. 7, pp. 1033–1039, 2002

2002
[158]

Stability Analysis of Delay Neural Networks With Impulsive Effects,

Z. Yang and D. Xu, “Stability Analysis of Delay Neural Networks With Impulsive Effects,”IEEE Trans. Circuits Syst. II: Express Briefs, vol. 52, no. 8, pp. 517–521, 2005

2005
[159]

Stability analysis for stochastic Cohen- Grossberg neural networks with mixed time delays,

Z. Wang, Y . Liu, M. Liet al., “Stability analysis for stochastic Cohen- Grossberg neural networks with mixed time delays,”IEEE Trans. Neural Netw., vol. 17, no. 3, pp. 814–820, 2006

2006
[160]

Exponential stability of impulsive Cohen- Grossberg-type BAM neural networks with delays and diffusion terms,

L. Wan and Q. Zhou, “Exponential stability of impulsive Cohen- Grossberg-type BAM neural networks with delays and diffusion terms,” inProc. 6th IEEE Int. Conf. Nat. Comput. (ICNC)), vol. 1. IEEE, 2010, pp. 282–286

2010
[161]

A Comprehensive Review of Stability Analysis of Continuous-Time Recurrent Neural Networks,

H. Zhang, Z. Wang, and D. Liu, “A Comprehensive Review of Stability Analysis of Continuous-Time Recurrent Neural Networks,” IEEE Trans. Neural Netw. Learn. Syst., vol. 25, no. 7, pp. 1229–1262, 2014

2014
[162]

Standard representation and unified stability analysis for dynamic artificial neural network models,

K.-K. K. Kim, E. R. Patr ´on, and R. D. Braatz, “Standard representation and unified stability analysis for dynamic artificial neural network models,”Neural Netw., vol. 98, pp. 251–262, 2018

2018
[163]

Safety Verification and Robustness Analysis of Neural Networks via Quadratic Constraints and Semidefinite Programming,

M. Fazlyab, M. Morari, and G. J. Pappas, “Safety Verification and Robustness Analysis of Neural Networks via Quadratic Constraints and Semidefinite Programming,”IEEE Trans. Autom. Control, vol. 67, no. 1, pp. 1–15, 2020

2020
[164]

Artificial intelligence techniques for sta- bility analysis and control in smart grids: Methodologies, applications, challenges and future directions,

Z. Shi, W. Yao, Z. Liet al., “Artificial intelligence techniques for sta- bility analysis and control in smart grids: Methodologies, applications, challenges and future directions,”Appl. Energy, vol. 278, p. 115733, 2020

2020
[165]

Stability-Certified Reinforcement Learning: A Control-Theoretic Perspective,

M. Jin and J. Lavaei, “Stability-Certified Reinforcement Learning: A Control-Theoretic Perspective,”IEEE Access, vol. 8, pp. 229 086– 229 100, 2020

2020
[166]

Reach-SDP: Reachability Analysis of Closed-Loop Systems with Neural Network Controllers via Semidefinite Programming,

H. Hu, M. Fazlyab, M. Morariet al., “Reach-SDP: Reachability Analysis of Closed-Loop Systems with Neural Network Controllers via Semidefinite Programming,” inProc. 59th IEEE Conf. Decis. Control (CDC). IEEE, 2020, pp. 5929–5934

2020
[167]

Stability Analysis Using Quadratic Constraints for Systems With Neural Network Controllers,

H. Yin, P. Seiler, and M. Arcak, “Stability Analysis Using Quadratic Constraints for Systems With Neural Network Controllers,”IEEE Trans. Autom. Control, vol. 67, no. 4, pp. 1980–1987, 2021

1980
[168]

Trustworthy Artificial Intelligence: A Review,

D. Kaur, S. Uslu, K. J. Rittichieret al., “Trustworthy Artificial Intelligence: A Review,”ACM Comput. Surv. (CSUR), vol. 55, no. 2, pp. 1–38, 2022

2022
[169]

Fusing physics-based and deep learning models for prognostics,

M. A. Chao, C. Kulkarni, K. Goebelet al., “Fusing physics-based and deep learning models for prognostics,”Reliab. Eng. Syst. Saf., vol. 217, p. 107961, 2022

2022
[170]

MotionLM: Multi-Agent Motion Forecasting as Language Modeling,

A. Seff, B. Cera, D. Chenet al., “MotionLM: Multi-Agent Motion Forecasting as Language Modeling,” inProc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), 2023, pp. 8579–8590

2023
[171]

XAI-Explainable artificial intelligence,

D. Gunning, M. Stefik, J. Choiet al., “XAI-Explainable artificial intelligence,”Sci. Robot., vol. 4, no. 37, p. eaay7120, 2019

2019
[172]

Decoding the Black Box: Extracting Explainable Decision Boundary Approximations from Ma- chine Learning Models for Real Time Safety Assurance of the National Airspace,

A. Grushin, J. Nanda, A. Tyagiet al., “Decoding the Black Box: Extracting Explainable Decision Boundary Approximations from Ma- chine Learning Models for Real Time Safety Assurance of the National Airspace,” inAIAA Scitech Forum, 2019, p. 136

2019
[173]

A historical perspective of explainable Artificial Intelligence,

R. Confalonieri, L. Coba, B. Wagneret al., “A historical perspective of explainable Artificial Intelligence,”Wiley Interdiscip. Rev.: Data Min. Knowl. Discov., vol. 11, no. 1, p. e1391, 2021

2021
[174]

The seven tools of causal inference, with reflections on machine learning,

J. Pearl, “The seven tools of causal inference, with reflections on machine learning,”Commun. ACM, vol. 62, no. 3, pp. 54–60, 2019

2019
[175]

Causality for Machine Learning,

B. Sch ¨olkopf, “Causality for Machine Learning,” inProbabilistic and Causal Inference: The Works of Judea Pearl, 2022, pp. 765–804

2022
[176]

Counterfactual Normalization: Proac- tively Addressing Dataset Shift and Improving Reliability Using Causal Mechanisms,

A. Subbaswamy and S. Saria, “Counterfactual Normalization: Proac- tively Addressing Dataset Shift and Improving Reliability Using Causal Mechanisms,”arXiv preprint arXiv:1808.03253, 2018

2018 arXiv
[177]

Preventing Failures Due to Dataset Shift: Learning Predictive Models That Transport,

A. Subbaswamy, P. Schulam, and S. Saria, “Preventing Failures Due to Dataset Shift: Learning Predictive Models That Transport,” inProc. 22th Int. Conf. Artif. Intell. Statist. (AISTATS). PMLR, 2019, pp. 3118–3127

2019
[178]

Input Validation for Neural Net- works via Runtime Local Robustness Verification,

J. Liu, L. Chen, A. Mineet al., “Input Validation for Neural Net- works via Runtime Local Robustness Verification,”arXiv preprint arXiv:2002.03339, 2020

2002 arXiv
[179]

Adversarially Learned One-Class Classifier for Novelty Detection,

M. Sabokrou, M. Khalooei, M. Fathyet al., “Adversarially Learned One-Class Classifier for Novelty Detection,” inProc. IEEE/CVF Com- put. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2018, pp. 3379– 3388

2018
[180]

Runtime Monitoring Neuron Activation Patterns,

C.-H. Cheng, G. N ¨uhrenberg, and H. Yasuoka, “Runtime Monitoring Neuron Activation Patterns,” inProc. IEEE Des. Autom. Test Eur. (DATE), 2019, pp. 300–303

2019
[181]

Outside the Box: Abstraction-Based Monitoring of Neural Networks,

T. A. Henzinger, A. Lukina, and C. Schilling, “Outside the Box: Abstraction-Based Monitoring of Neural Networks,” inECAI 2020. IOS Press, 2020, pp. 2433–2440

2020
[182]

A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks,

D. Hendrycks and K. Gimpel, “A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks,” inProc. Int. Conf. Learn. Represent. (ICLR), 2017, pp. 1–9

2017
[183]

Enhancing The Reliability of Out-of- distribution Image Detection in Neural Networks,

S. Liang, Y . Li, and R. Srikant, “Enhancing The Reliability of Out-of- distribution Image Detection in Neural Networks,” inProc. Int. Conf. Learn. Represent. (ICLR), 2017, pp. 1–27

2017
[184]

From Ordinary Differential Equations to Structural Causal Models: the deterministic case,

J. M. Mooij, D. Janzing, and B. Sch ¨olkopf, “From Ordinary Differential Equations to Structural Causal Models: the deterministic case,” inProc. 29th Conf. Uncert. Artif. Intell. (UAI). AUAI Press, 2013, pp. 440– 448

2013
[185]

Models for Interacting Populations,

J. Murray, “Models for Interacting Populations,”Mathematical Biol- ogy: I. An Introduction, pp. 79–118, 2002

2002
[186]

Beyond Finite Layer Neural Networks: Bridging Deep Architectures and Numerical Differential Equations,

Y . Lu, A. Zhong, Q. Liet al., “Beyond Finite Layer Neural Networks: Bridging Deep Architectures and Numerical Differential Equations,” inProc. 35th Int. Conf. Mach. Learn. (ICML). PMLR, 2018, pp. 3276–3285

2018
[187]

Stable architectures for deep neural networks,

E. Haber and L. Ruthotto, “Stable architectures for deep neural networks,”Inverse Probl., vol. 34, no. 1, p. 014004, 2017

2017
[188]

Vehicle Dynamics,

D. Schramm, M. Hiller, and R. Bardini, “Vehicle Dynamics,”Modeling and Simulation. Berlin, Heidelberg, vol. 151, 2014

2014
[189]

Neural Autoregressive Flows,

C.-W. Huang, D. Krueger, A. Lacosteet al., “Neural Autoregressive Flows,” inProc. 35th Int. Conf. Mach. Learn. (ICML). PMLR, 2018, pp. 2078–2087

2018
[190]

Fast Decoding in Sequence Models Using Discrete Latent Variables,

L. Kaiser, S. Bengio, A. Royet al., “Fast Decoding in Sequence Models Using Discrete Latent Variables,” inProc. 35th Int. Conf. Mach. Learn. (ICML). PMLR, 2018, pp. 2390–2399

2018
[191]

Probabilistic Interpretations of Recurrent Neural Networks,

Y . J. Choe, J. Shin, and N. Spencer, “Probabilistic Interpretations of Recurrent Neural Networks,”Probabilistic Graphical Models, 2017

2017
[192]

Recognizing recurrent neural networks (rRNN): Bayesian inference for recurrent neural networks,

S. Bitzer and S. J. Kiebel, “Recognizing recurrent neural networks (rRNN): Bayesian inference for recurrent neural networks,”Biol. Cybern., vol. 106, pp. 201–217, 2012

2012
[193]

Hidden Markov models,

S. R. Eddy, “Hidden Markov models,”Curr. Opin. Struct. Biol., vol. 6, no. 3, pp. 361–365, 1996

1996
[194]

Statistical Inference for Probabilistic Functions of Finite State Markov Chains,

L. E. Baum and T. Petrie, “Statistical Inference for Probabilistic Functions of Finite State Markov Chains,”Ann. Math. Stat., vol. 37, no. 6, pp. 1554–1563, 1966

1966
[195]

A. H. Jazwinski,Stochastic Processes and Filtering Theory. Courier Corporation, 2007

2007
[196]

Parameter estimation in general state- space models using particle methods,

A. Doucet and V . B. Tadi ´c, “Parameter estimation in general state- space models using particle methods,”Ann. Inst. Stat. Math., vol. 55, pp. 409–422, 2003

2003
[197]

Particle filtering,

P. M. Djuric, J. H. Kotecha, J. Zhanget al., “Particle filtering,”IEEE Signal Process. Mag., vol. 20, no. 5, pp. 19–38, 2003

2003
[198]

Neural Turing Machines,

A. Graves, G. Wayne, and I. Danihelka, “Neural Turing Machines,” arXiv preprint arXiv:1410.5401, 2014

2014 arXiv
[199]

Hybrid computing using a neural network with dynamic external memory,

A. Graves, G. Wayne, M. Reynoldset al., “Hybrid computing using a neural network with dynamic external memory,”Nature, vol. 538, no. 7626, pp. 471–476, 2016

2016
[200]

Neural Attention Memory,

H. Nam and S. B. Seo, “Neural Attention Memory,”arXiv preprint arXiv:2302.09422, 2023

2023 arXiv
[201]

Memory Networks,

J. Weston, S. Chopra, and A. Bordes, “Memory Networks,”arXiv preprint arXiv:1410.3916, 2014

2014 arXiv
[202]

Neurons with graded response have collective compu- tational properties like those of two-state neurons

J. J. Hopfield, “Neurons with graded response have collective compu- tational properties like those of two-state neurons.”Proc. Natl. Acad. Sci. U. S. A., vol. 81, no. 10, pp. 3088–3092, 1984

1984
[203]

Dense Associative Memory for Pattern Recognition,

D. Krotov and J. J. Hopfield, “Dense Associative Memory for Pattern Recognition,” inProc. 29th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2016, pp. 1–9

2016
[204]

Attention is All you Need,

A. Vaswani, N. Shazeer, N. Parmaret al., “Attention is All you Need,” inProc. 30th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2017

2017
[205]

Neural networks and physical systems with emergent collective computational abilities

J. J. Hopfield, “Neural networks and physical systems with emergent collective computational abilities.”Proc. Natl. Acad. Sci. U. S. A., vol. 79, no. 8, pp. 2554–2558, 1982

1982
[206]

Social LSTM: Human Trajectory Prediction in Crowded Spaces,

A. Alahi, K. Goel, V . Ramanathanet al., “Social LSTM: Human Trajectory Prediction in Crowded Spaces,” inProc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2016, pp. 961–971

2016
[207]

Deep Imitative Models for Flexible Inference, Planning, and Control,

N. Rhinehart, R. McAllister, and S. Levine, “Deep Imitative Models for Flexible Inference, Planning, and Control,” inInt. Conf. Learn. Represent. (ICLR), 2020, pp. 1–20

2020
[208]

Model-Based Imitation Learning for Urban Driving,

A. Hu, G. Corrado, N. Griffithset al., “Model-Based Imitation Learning for Urban Driving,” inProc. 35th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2022, pp. 20 703–20 716

2022
[209]

Conditional Neural Processes,

M. Garnelo, D. Rosenbaum, C. Maddisonet al., “Conditional Neural Processes,” inProc. 35th Int. Conf. Mach. Learn. (ICML). PMLR, 2018, pp. 1704–1713

2018
[210]

Benchmarking Safety Monitors for Image Classifiers with Machine Learning,

R. S. Ferreira, J. Arlat, J. Guiochetet al., “Benchmarking Safety Monitors for Image Classifiers with Machine Learning,” inProc. IEEE Pac. Rim Int. Symp. Dependable Comput. (PRDC), 2021, pp. 7–16

2021
[211]

End-to-End Object Detec- tion with Transformers,

N. Carion, F. Massa, G. Synnaeveet al., “End-to-End Object Detec- tion with Transformers,” inProc. Eur. Conf. Comput. Vis. (ECCV). Springer, 2020, pp. 213–229

2020
[212]

Motion Transformer with Global Intention Localization and Local Movement Refinement,

S. Shi, L. Jiang, D. Daiet al., “Motion Transformer with Global Intention Localization and Local Movement Refinement,” inProc. 36th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2022, pp. 6531–6543

2022
[213]

V ADv2: End-to-End Vector- ized Autonomous Driving via Probabilistic Planning,

S. Chen, B. Jiang, H. Gaoet al., “V ADv2: End-to-End Vector- ized Autonomous Driving via Probabilistic Planning,”arXiv preprint arXiv:2402.13243, 2024

2024 arXiv
[214]

The perceptron: A probabilistic model for information storage and organization in the brain

F. Rosenblatt, “The perceptron: A probabilistic model for information storage and organization in the brain.”Psychol. Rev., vol. 65, no. 6, p. 386, 1958

1958
[215]

You Only Look Once: Unified, Real-Time Object Detection,

J. Redmon, S. Divvala, R. Girshicket al., “You Only Look Once: Unified, Real-Time Object Detection,” inProc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2016, pp. 779–788

2016
[216]

Fast R-CNN,

R. Girshick, “Fast R-CNN,” inProc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), 2015, pp. 1440–1448

2015
[217]

Image Segmenta- tion Using K-means Clustering Algorithm and Subtractive Clustering Algorithm,

N. Dhanachandra, K. Manglem, and Y . J. Chanu, “Image Segmenta- tion Using K-means Clustering Algorithm and Subtractive Clustering Algorithm,”Procedia Computer Science, vol. 54, pp. 764–771, 2015

2015
[218]

Lorenz,Die R ¨uckseite des Spiegels

K. Lorenz,Die R ¨uckseite des Spiegels. Versuch einer Naturgeschichte menschlichen Erkennens. M ¨unchen and Z ¨urich: Piper & Co Verlag, 1973

1973
[219]

Deformable DETR: Deformable Transformers for End-to-End Object Detection,

X. Zhu, W. Su, L. Luet al., “Deformable DETR: Deformable Transformers for End-to-End Object Detection,”arXiv preprint arXiv:2010.04159, 2020

2010 arXiv
[220]

Universal adversarial perturbations against object detection,

D. Li, J. Zhang, and K. Huang, “Universal adversarial perturbations against object detection,”Pattern Recognit., vol. 110, p. 107584, 2021

2021
[221]

Humanlike Driving: Empirical Decision- Making System for Autonomous Vehicles,

L. Li, K. Ota, and M. Dong, “Humanlike Driving: Empirical Decision- Making System for Autonomous Vehicles,”IEEE Trans. Veh. Technol., vol. 67, no. 8, pp. 6814–6823, 2018

2018
[222]

Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction,

S. Narayanan, R. Moslemi, F. Pittalugaet al., “Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction,” inProc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2021, pp. 15 799–15 808

2021
[223]

Learning by Cheating,

D. Chen, B. Zhou, V . Koltunet al., “Learning by Cheating,” inProc. Conf. Robo. Learn. (CoRL). PMLR, 2020, pp. 66–75

2020
[224]

Lyapunov Stability Theory of Nonsmooth Systems,

D. Shevitz and B. Paden, “Lyapunov Stability Theory of Nonsmooth Systems,”IEEE Trans. Autom. Control, vol. 39, no. 9, pp. 1910–1914, 1994

1910
[225]

Lyapunov Stability Theory,

S. Sastry and S. Sastry, “Lyapunov Stability Theory,”Nonlinear Sys- tems: Analysis, Stability, and Control, pp. 182–234, 1999

1999
[226]

Dissipative deep neural dynamical systems,

J. Drgo ˇna, A. Tuor, S. Vasishtet al., “Dissipative deep neural dynamical systems,”IEEE Open J. of Control Syst., vol. 1, pp. 100–112, 2022

2022
[227]

On Information and Sufficiency,

S. Kullback and R. A. Leibler, “On Information and Sufficiency,”Ann. Math. Stat., vol. 22, no. 1, pp. 79–86, 1951

1951
[228]

Online Multi-Target Tracking Using Recurrent Neural Networks,

A. Milan, S. H. Rezatofighi, A. Dicket al., “Online Multi-Target Tracking Using Recurrent Neural Networks,” inProc. AAAI Conf. Artif. Intell. (AAAI), vol. 31, no. 1, 2017

2017
[229]

Motion Forecasting in Continuous Driving,

N. Song, B. Zhang, X. Zhuet al., “Motion Forecasting in Continuous Driving,”arXiv preprint arXiv:2410.06007, 2024

2024 arXiv
[230]

PlanT: Explainable Planning Transformers via Object-Level Representations,

K. Renz, K. Chitta, O.-B. Merceaet al., “PlanT: Explainable Planning Transformers via Object-Level Representations,” inProc. Conf. Robo. Learn. (CoRL). PMLR, 2022, pp. 459–470

2022
[231]

European Parliament, “Corrigendum to the position of the European Parliament adopted at first reading on 13 March 2024 with a view to the adoption of Regulation (EU) 2024/... of the European Parliament and of the Council laying down harmonised rules on artificial intelligence ...

2024
[232]

(2021, July 9)Call for Public Comments on ”AI Governance Guidelines for Implementation of AI Principles Ver

Ministry of Economy, Trade and Industry (METI). (2021, July 9)Call for Public Comments on ”AI Governance Guidelines for Implementation of AI Principles Ver. 1.0” Opens. [Online]. Available: https://www.meti.go.jp/english/press/2021/0709 004.html

2021
[233]

Positionspapier: Ein Rechtsrahmen f ¨ur K ¨unstliche Intelligenz,

F. Thouvenin, M. Christen, A. Bernsteinet al., “Positionspapier: Ein Rechtsrahmen f ¨ur K ¨unstliche Intelligenz,”Digital Society Initiative, 2021

2021
[234]

(2022, June)Bill C-27, Consumer Privacy Protection Act, PART 3 Artificial Intelligence and Data Act

House of Commons of Canada, First Session. (2022, June)Bill C-27, Consumer Privacy Protection Act, PART 3 Artificial Intelligence and Data Act. [Online]. Available: https://www.parl.ca/DocumentViewer/e n/44-1/bill/C-27/first-reading

2022
[235]

(2019, June)Developing Responsible Artificial Intelligence: Release of the New Generation of Artificial Intelligence Governance Principles

National Committee on the Governance of the New Generation of Artificial Intelligence. (2019, June)Developing Responsible Artificial Intelligence: Release of the New Generation of Artificial Intelligence Governance Principles. [Online]. Available: https: //www.most.gov.cn/kjbg...

2019
[236]

Artificial Intelligence Risk Management Framework (AI RMF 1.0),

E. Tabassi, “Artificial Intelligence Risk Management Framework (AI RMF 1.0),”National Institute of Standards and Technology (NIST), Gaithersburg, MD, 2023, DOI: 10.6028/NIST.AI.100-1

2023 doi
[237]

Mind the gaps: Assuring the safety of autonomous systems from an engineering, ethical, and legal perspective,

S. Burton, I. Habli, T. Lawtonet al., “Mind the gaps: Assuring the safety of autonomous systems from an engineering, ethical, and legal perspective,”Artif. Intell., vol. 279, pp. 1–16, 2020, Art. no. 103201

2020
[238]

Closing the gaps: Complexity and uncertainty in the safety assurance and regulation of automated driving,

S. Burton and J. A. McDermid, “Closing the gaps: Complexity and uncertainty in the safety assurance and regulation of automated driving,” 2023. [Online]. Available: https://publica-rest.fraunhofer.de/ server/api/core/bitstreams/c0198205-8061-4fcf-bfa3-02e37bc2780c/co ntent

2023
[239]

Energy-based Out-of-distribution Detection,

W. Liu, X. Wang, J. Owenset al., “Energy-based Out-of-distribution Detection,” inProc. 33th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2020, pp. 21 464–21 475

2020
[240]

Likelihood Ratios and Generative Classifiers for Unsupervised Out-of-Domain Detection in Task Oriented Dialog,

V . Gangal, A. Arora, A. Einolghozatiet al., “Likelihood Ratios and Generative Classifiers for Unsupervised Out-of-Domain Detection in Task Oriented Dialog,” inProc. AAAI Conf. on Artif. Intell., vol. 34, no. 05, 2020, pp. 7764–7771

2020
[241]

A Deep Generative Distance-Based Classifier for Out-of-Domain Detection with Mahalanobis Space,

H. Xu, K. He, Y . Yanet al., “A Deep Generative Distance-Based Classifier for Out-of-Domain Detection with Mahalanobis Space,” in Proc. 28th Int. Conf. Comput. Linguist., 2020, pp. 1452–1460

2020
[242]

Insufficiency-driven DNN error detection in the context of SOTIF on traffic sign recognition use case,

L. Hacker and J. Seewig, “Insufficiency-driven DNN error detection in the context of SOTIF on traffic sign recognition use case,”IEEE Open J. Intell. Transp. Syst., vol. 4, pp. 58–70, 2023

2023
[243]

Detection of False Positive and False Negative Samples in Semantic Segmentation,

M. Rottmann, K. Maag, R. Chanet al., “Detection of False Positive and False Negative Samples in Semantic Segmentation,” inProc. IEEE Des. Autom. Test Eur. (DATE), 2020, pp. 1351–1356

2020
[244]

Addressing uncertainty in the safety assurance of machine-learning,

S. Burton and B. Herd, “Addressing uncertainty in the safety assurance of machine-learning,”Front. Comput. Sci., vol. 5, pp. 1–17, 2023, Art. no. 1132580

2023
[245]

Identifying Unknown Unknowns in the Open World: Representations and Policies for Guided Exploration,

H. Lakkaraju, E. Kamar, R. Caruanaet al., “Identifying Unknown Unknowns in the Open World: Representations and Policies for Guided Exploration,” inProc. AAAI Conf. on Artif. Intell., vol. 31, no. 1, 2017

2017
[246]

Achieving Robustness in the Wild via Adversarial Mixing With Disentangled Representations,

S. Gowal, C. Qin, P.-S. Huanget al., “Achieving Robustness in the Wild via Adversarial Mixing With Disentangled Representations,” in Proc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2020, pp. 1211–1220

2020
[247]

Causal Graph Attention Network with Disentangled Representations for Complex Systems Fault Detection,

J. Liu, S. Zheng, and C. Wang, “Causal Graph Attention Network with Disentangled Representations for Complex Systems Fault Detection,” Reliab. Eng. Syst. Saf., vol. 235, p. 109232, 2023

2023
[248]

Controlled generation of unseen faults for Partial and Open-Partial domain adaptation,

K. Rombach, G. Michau, and O. Fink, “Controlled generation of unseen faults for Partial and Open-Partial domain adaptation,”Reliab. Eng. Syst. Saf., vol. 230, p. 108857, 2023

2023
[249]

Kahneman,Thinking, Fast and Slow

D. Kahneman,Thinking, Fast and Slow. New York, NY , USA: Farrar, Straus and Giroux, 2011

2011
[250]

Tutorial overview of model predictive control,

J. B. Rawlings, “Tutorial overview of model predictive control,”IEEE control systems magazine, vol. 20, no. 3, pp. 38–52, 2000

2000
[251]

A software framework for embedded nonlinear model predictive control using a gradient-based augmented Lagrangian approach (GRAMPC),

T. Englert, A. V ¨olz, F. Mesmeret al., “A software framework for embedded nonlinear model predictive control using a gradient-based augmented Lagrangian approach (GRAMPC),”Optim. Eng., vol. 20, pp. 769–809, 2019

2019
[252]

Moving Horizon Estimation,

D. A. Allan and J. B. Rawlings, “Moving Horizon Estimation,” Handbook of Model Predictive Control, pp. 99–124, 2019

2019
[253]

Argoverse: 3D Tracking and Forecasting With Rich Maps,

M.-F. Chang, J. Lambert, P. Sangkloyet al., “Argoverse: 3D Tracking and Forecasting With Rich Maps,” inProc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2019, pp. 8748–8757

2019
[254]

INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps,

W. Zhan, L. Sun, D. Wanget al., “INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps,”arXiv preprint arXiv:1910.03088, 2019

1910 arXiv
[255]

One Thousand and One Hours: Self-driving Motion Prediction Dataset,

J. Houston, G. Zuidhof, L. Bergaminiet al., “One Thousand and One Hours: Self-driving Motion Prediction Dataset,” inInt. Conf. Learn. Represent. (ICLR). PMLR, 2021, pp. 409–418

2021
[256]

NuPlan: A closed-loop ML- based planning benchmark for autonomous vehicles,

H. Caesar, J. Kabzan, K. S. Tanet al., “NuPlan: A closed-loop ML- based planning benchmark for autonomous vehicles,”arXiv preprint arXiv:2106.11810, 2021

2021 arXiv
[257]

Enhancing System Self- Awareness and Trust of AI: A Case Study in Trajectory Prediction and Planning,

L. Ullrich, Z. Mujirishvili, and K. Graichen, “Enhancing System Self- Awareness and Trust of AI: A Case Study in Trajectory Prediction and Planning,” inProc. IEEE Intell. Veh. Symp. (IV), 2025, pp. 1–8, accepted

2025
[258]

MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Features,

A. Bardes, J. Ponce, and Y . LeCun, “MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Features,”arXiv preprint arXiv:2307.12698, 2023

2023 arXiv
[259]

Bridging Past and Future: End- to-End Autonomous Driving with Historical Prediction and Planning,

B. Zhang, N. Song, X. Jinet al., “Bridging Past and Future: End- to-End Autonomous Driving with Historical Prediction and Planning,” arXiv preprint arXiv:2503.14182, 2025

2025 arXiv
[260]

AUTOtech.agil: Ar- chitecture and Technologies for Orchestrating Automotive Agility,

R. van Kempen, B. Lampe, M. Leuffenet al., “AUTOtech.agil: Ar- chitecture and Technologies for Orchestrating Automotive Agility,” in Proc. of 32nd Aachen Colloquium Sustainable Mobility, 2023

2023
[261]

Invariant Models for Causal Transfer Learning,

M. Rojas-Carulla, B. Sch ¨olkopf, R. Turneret al., “Invariant Models for Causal Transfer Learning,”J. Mach. Learn. Res., vol. 19, no. 36, pp. 1–34, 2018

2018
[262]

Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distribu- tions,

S. Magliacane, T. Van Ommen, T. Claassenet al., “Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distribu- tions,” inProc. 31st Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2018, pp. 1–11

2018
[263]

Estimating the Support of a High-Dimensional Distribution,

B. Sch ¨olkopf, J. C. Platt, J. Shawe-Tayloret al., “Estimating the Support of a High-Dimensional Distribution,”Neural Comput., vol. 13, no. 7, pp. 1443–1471, 2001

2001
[264]

Towards Open Set Deep Networks,

A. Bendale and T. E. Boult, “Towards Open Set Deep Networks,” in Proc. IEEE/CVF Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), 2016, pp. 1563–1572

2016
[265]

Decision-theoretic troubleshooting,

D. Heckerman, J. S. Breese, and K. Rommelse, “Decision-theoretic troubleshooting,”Commun. ACM, vol. 38, no. 3, pp. 49–57, 1995

1995
[266]

Towards accountable ai: Hybrid human-machine analyses for characterizing system failure,

B. Nushi, E. Kamar, and E. Horvitz, “Towards accountable ai: Hybrid human-machine analyses for characterizing system failure,” inProc. AAAI Conf. Human Comput. Crowdsourc. (HCOMP), vol. 6, 2018, pp. 126–135

2018
[267]

Benchmarking Neural Network Robustness to Common Corruptions and Perturbations,

D. Hendrycks and T. Dietterich, “Benchmarking Neural Network Robustness to Common Corruptions and Perturbations,” inInt. Conf. Learn. Represent. (ICLR), 2019, pp. 1–16

2019
[268]

Learning models with uniform per- formance via distributionally robust optimization,

J. C. Duchi and H. Namkoong, “Learning models with uniform per- formance via distributionally robust optimization,”Ann. Stat., vol. 49, no. 3, pp. 1378–1406, 2021

2021
[269]

Causality from a Distributional Robustness Point of View,

N. Meinshausen, “Causality from a Distributional Robustness Point of View,” inIEEE Data Science Workshop (DSW), 2018, pp. 6–10

2018
[270]

The s-value: evaluating stability with respect to distributional shifts,

S. Gupta and D. Rothenh ¨ausler, “The s-value: evaluating stability with respect to distributional shifts,” inProc. 36th Int. Conf. Neural Inf. Process. Syst. (NeurIPS), 2024, pp. 1–13

2024
[271]

Mandoline: Model Evaluation under Distribution Shift,

M. Chen, K. Goel, N. S. Sohoniet al., “Mandoline: Model Evaluation under Distribution Shift,” inProc. 38th Int. Conf. Mach. Learn. (ICML). PMLR, 2021, pp. 1617–1629

2021
[272]

A unifying causal frame- work for analyzing dataset shift-stable learning algorithms,

A. Subbaswamy, B. Chen, and S. Saria, “A unifying causal frame- work for analyzing dataset shift-stable learning algorithms,”J. Causal Inference, vol. 10, no. 1, pp. 64–89, 2022. Lars Ullrichreceived the M.Sc. degree in mecha- tronics from Friedrich–Alexander–Universita ¨at Er...

2022
[2024]

Available: https://www.nist.gov/system/files/documen ts/2024/05/21/AISI-vision-21May2024.pdf

[Online]. Available: https://www.nist.gov/system/files/documen ts/2024/05/21/AISI-vision-21May2024.pdf

2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Object Classification Using CNN-Based Fusion of Vision and LIDAR in Autonomous Vehicle Environment,

H. Gao, B. Cheng, J. Wanget al., “Object Classification Using CNN-Based Fusion of Vision and LIDAR in Autonomous Vehicle Environment,”IEEE Trans. Ind. Inform., vol. 14, no. 9, pp. 4224–4231, 2018

2018

[2] [2]

Toward Explainable Artificial Intelligence for Regression Models: A methodological perspective,

S. Letzgus, P. Wagner, J. Ledereret al., “Toward Explainable Artificial Intelligence for Regression Models: A methodological perspective,” IEEE Signal Process. Mag., vol. 39, no. 4, pp. 40–58, 2022

2022

[3] [3]

A comprehensive survey of clustering algorithms: State-of-the-art machine learning ap- plications, taxonomy, challenges, and future research prospects,

A. E. Ezugwu, A. M. Ikotun, O. O. Oyeladeet al., “A comprehensive survey of clustering algorithms: State-of-the-art machine learning ap- plications, taxonomy, challenges, and future research prospects,”Eng. Appl. Artif. Intell., vol. 110, pp. 1–43, 2022, Art. no. 104743

2022

[4] [4]

Time-series forecasting with deep learning: a survey,

B. Lim and S. Zohren, “Time-series forecasting with deep learning: a survey,”Phil. Trans. R. Soc., vol. 379, no. 2194, pp. 1–14, 2021, Art. no. 20200209

2021

[5] [5]

Application of artificial intelligence algo- rithms in image processing,

X. Zhang and W. Dahu, “Application of artificial intelligence algo- rithms in image processing,”J. Vis. Commun. Image Represent., vol. 61, pp. 42–49, 2019

2019

[6] [6]

An introduction to Deep Learning in Natural Language Processing: Models, techniques, and tools,

I. Lauriola, A. Lavelli, and F. Aiolli, “An introduction to Deep Learning in Natural Language Processing: Models, techniques, and tools,”Neurocomputing, vol. 470, pp. 443–456, 2022

2022

[7] [7]

Causal inference and the data-fusion problem,

E. Bareinboim and J. Pearl, “Causal inference and the data-fusion problem,”Proc. Natl. Acad. Sci., vol. 113, no. 27, pp. 7345–7352, 2016

2016

[8] [8]

A Comprehensive Survey on Transfer Learning,

F. Zhuang, Z. Qi, K. Duanet al., “A Comprehensive Survey on Transfer Learning,”Proc. IEEE, vol. 109, no. 1, pp. 43–76, 2020

2020