Pith. sign in

REVIEW 5 major objections 5 minor 33 references

ANSR-DT: A Neuro-Symbolic Framework for Adaptive and Explainable Digital Twins

T0 review · 5 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read ANSR-DT couples neural detection, symbolic rules, and reinforcement learning in one digital twin pipeline, reporting 99.5% peak accuracy on synthetic industrial events.

desk verdict A plausible neuro-symbolic digital twin pipeline, but the reported results are internally inconsistent and the abstract overstates what the body actually evaluates. read the letter →

arxiv 2501.08561 v5 pith:BYWEOS7A submitted 2025-01-15 cs.AI cs.HCcs.LGcs.SC

Safayat Bin Hakim , Muhammad Adil , Alvaro Velasquez , Houbing Herbert Song This is my paper · ORCID
classification cs.AIcs.HCcs.LGcs.SC
keywords digitaltwinneuro-symbolicAIanomalydetectionCNN-LSTMsymbolicruleextractionreinforcementlearningPPOexplainable
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

ANSR-DT is a proposed architecture for digital twins that closes the loop between three jobs usually done separately: neural pattern detection, symbolic explanation, and reinforcement-learning control. The paper claims that this single pipeline detects dynamic events in multivariate industrial sensor streams with up to 99.5% peak validation accuracy, extracts a stable set of 14 human-readable rules, and improves its control policy through extended PPO training while the explanations remain unchanged. If the claim holds, industrial operators would get an AI monitor that says not just 'anomaly' but why, and that adapts its response over time without losing that transparency. The reported evidence comes from a synthetic 5,000-sample industrial dataset, and the authors position the framework as a foundation for scalable, explainable digital twins in safety-critical settings.

What carries the argument

The load-bearing mechanism is the confidence-filtered rule loop. Neural outputs become candidate rules; a candidate survives only if $\operatorname{conf}(r)=\frac{\operatorname{support}(r\Rightarrow c)}{\operatorname{support}(r)}\cdot\frac{TP}{TP+FP}$ exceeds $\tau=0.85$. Surviving rules enter the symbolic knowledge base, and the PPO agent's clipped surrogate objective drives policy updates, with a rule-update operation closing the loop. This is what lets the same system detect patterns, explain them, and adapt without full retraining.

What would settle it

A reader could settle the transfer claim by running the released pipeline on the anomaly benchmark referenced in the abstract and comparing detection F1, precision, recall, and rule stability against the synthetic results; the manuscript body as provided reports no such benchmark numbers, so that comparison is the decisive missing experiment.

Watch

Extended reading notes

Core claim

The paper's central claim, on its own terms, is that interpretability and adaptation can be coupled rather than traded off. A CNN-LSTM with attention extracts temporal patterns from multivariate sensor data; a symbolic reasoner converts those patterns into a stable set of human-readable rules, 14 in the reported experiments with most above 0.9 confidence; and a PPO agent uses the resulting state representation to choose control actions, with the rule base and policy updated together. On the synthetic industrial dataset the system reaches 99.5% peak validation accuracy, improves recall by 17.82% over the CNN-LSTM baseline, and extended PPO training raises explained variance from 0.447 to 0.547 while the extracted symbolic rules remain unchanged.

Load-bearing premise

The load-bearing premise is that a synthetic dataset of 5,000 samples with 5% injected events, event correlations of 0.3 to 0.5, and Savitzky-Golay smoothing resembles real industrial sensor conditions closely enough that performance and rule counts transfer; if real operational noise and event patterns differ, the reported accuracy and adaptation gains may not survive.

Editorial extensions

If this is right

  • A digital twin built on this pipeline would produce a human-readable explanation for each detected deviation, with the explanation remaining consistent even while the control policy is retrained.
  • Operators would see fewer missed critical events: the framework reports a 17.82% recall improvement over the CNN-LSTM baseline, at the cost of more false positives that humans can review.
  • Longer reinforcement-learning training would make the system's control actions more decisive without destabilizing the symbolic explanation layer, as shown by explained variance rising from 0.447 to 0.547 while rules stay fixed at 14.
  • Removing the symbolic component measurably hurts detection (7.4% F1 drop), so interpretability is not an add-on in this design; the rule loop actively guides the neural attention mechanism.
  • The framework's reported 99.5% peak accuracy and rule stability position it as a template for trustworthy digital twins in safety-critical monitoring, pending validation on real operational data.

Reading between the lines

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

  • Editorial inference: the stability of the 14 rules across policy updates implies the explanation layer could serve as an audit trail, letting an operator or regulator reconstruct why an action was taken by querying a small logic program rather than a neural network.
  • Editorial inference: the reported 35% overhead and degradation beyond roughly 50 rules suggest a concrete stress test—whether hierarchical rule modules or incremental re-validation restores inference speed on larger rule bases; the paper names these as future work but does not test them.
  • Editorial inference: because the system trades 14.50% precision for a 17.82% recall gain, the right deployment question is whether false alarms are cheaper than missed events; the paper does not measure that cost trade-off, but its own numbers make the question unavoidable.
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.

Referee Report

5 major / 5 minor

Summary. The paper proposes ANSR-DT, a three-layer neuro-symbolic framework for industrial digital twins that combines a CNN-LSTM detector with attention, ProbLog-based symbolic rule extraction, and a PPO reinforcement-learning adaptation layer. On a synthetic 5,000-sample industrial dataset it reports 99.5% peak validation accuracy, 14 stable extracted rules, and an increase in PPO explained variance from 0.447 to 0.547. The abstract additionally claims comparison against eight baselines and validation on the Skoltech Anomaly Benchmark (SKAB).

Significance. Taken at face value, the proposed integration is timely and the open-source release is a useful contribution to reproducibility. However, the current evidence does not support the headline claims: the quantitative evaluation is internally contradictory, the abstract advertises experiments that do not appear in the body, and the described training configuration is inconsistent with the stated class-imbalance weighting. The conceptual contribution of the framework cannot be properly assessed until these issues are resolved.

major comments (5)
  1. [V-B (Table II and Fig. 6)] The two reported evaluations of the same ANSR-DT model cannot both be correct. Table II reports precision 0.80±0.05, recall 0.75±0.04, F1 0.77, and ROC-AUC 0.85±0.03, while the confusion matrix in Fig. 6a (343 TN, 0 FP, 153 TP, 3 FN) yields precision 1.00, recall 0.981, F1 0.990, and accuracy 0.994, and Fig. 6b reports AP=1.00. No thresholding or averaging protocol described in Section V-A can reconcile these values. The central claim of competitive predictive performance is therefore unsupported by an internally consistent measurement.
  2. [V-A and Fig. 6a] The confusion matrix total is inconsistent with the reported dataset split. A 60/20/20 split of 5,000 samples gives a 1,000-sample test set, and the 5% event injection rate gives about 50 positive test samples; the matrix instead has 499 total samples and 156 labeled positive (153 TP + 3 FN). The figure therefore does not correspond to the evaluation described in Section V-A, further undermining the reported performance numbers.
  3. [Abstract vs. Sections IV-VII] The abstract promises “experiments against eight baselines” and “additional validation on the Skoltech Anomaly Benchmark (SKAB),” but the body describes only a single CNN-LSTM baseline and contains no SKAB results anywhere in Sections IV through VII. These claims are central to the paper’s assertions of competitiveness and transferability; they must be substantiated with actual experiments or removed from the abstract.
  4. [V-A (class weighting)] The reported class weights (0: 64.7, 1: 0.5) are the reverse of what is needed to emphasize the minority key-event class: class 0 (normal, roughly 95% of samples) receives the large weight. The text states that this weighting “ensur[es] that the minority class (key events) received appropriate emphasis,” so the described training configuration is internally inconsistent and the claimed recall improvement cannot be attributed to the stated protocol.
  5. [IV-B2 and V-C (rule extraction)] The 14 rules are “extracted from test predictions” and then reported as stable “across PPO training durations” on the same condensed dataset. This is an internal consistency check rather than an independent validation, and it does not support the abstract’s general claim of “stable rule extraction.” The authors should validate the extracted rules on held-out data or clearly describe a training-data extraction procedure before claiming stability.
minor comments (5)
  1. [III-C2 (Eq. 4)] The rule confidence formula multiplies P(c|r) by rule precision; the text should define the populations over which support(r), support(r⇒c), TP, and FP are computed, since the two factors may be redundant or require different denominators.
  2. [V-B (Fig. 6b)] An “AP = 1.00” value is presented as a precision-recall curve, but no threshold sweep or operating-point sampling is described; a single confusion matrix cannot by itself produce a precision-recall curve.
  3. [IV-F and References] The text refers to ProbLog 2.2 while reference [27] is titled “Problog 2”; please align the name and version.
  4. [IV-A and V-A] The data are described both as “5,000 samples collected at 5-minute intervals” and as “5,000 multivariate time series sequences”; please clarify whether the CNN-LSTM input is a single sample, a sliding window, or a full sequence, and how the walk-forward cross-validation is applied in each case.
  5. [Fig. 8 caption] The caption states that the legend is omitted for brevity; since the interpretability claim relies on this figure, please include a legend or a fully self-contained description of the node and edge types.

Circularity Check

1 steps flagged · score 4.0 of 10

Rule-extraction 'reliability' is an internal consistency check on the same test predictions used to generate the rules; the otherwise independent predictive claims are additionally clouded by contradictory reported metrics and missing promised external validation.

  1. fitted input called prediction [Section IV-B.2, Eq. (4); Section V-C]
    "The framework extracted 14 rules from test predictions, with the rule count stabilizing at 14 across PPO training durations of 10,240 and 200,704 timesteps. This consistency, despite varying RL policies, demonstrates the symbolic component’s reliability in translating neural predictions into interpretable rules on the condensed dataset. ... For each candidate rule, we calculated confidence scores using the formula shown in Eq. 4 ... Rules with confidence scores above threshold τ=0.85 were incorporated into the knowledge base ..."

    The 14 rules are extracted from the model's own test predictions, and Eq. (4) evaluates each candidate rule's confidence on the same test predictions and labels, with rules below τ=0.85 discarded before reporting. Reporting that surviving rules have confidence >0.9 and calling this 'high reliability' is therefore a statement about the internal selection criterion, not an independent validation. The 'stability' of the rule count across PPO runs is likewise measured on the same extracted-rule/test-prediction pipeline, so it does not provide independent evidence that the symbolic component generalizes.

full rationale

No load-bearing self-citation chain or imported uniqueness theorem exists here; references [22] and [25] include co-authors but are used for general background, not to force the framework's design. The central classification claim (99.5% peak validation accuracy, recall gain over CNN-LSTM) is an independent empirical measurement, not derived from its own inputs. However, the paper's rule-extraction contribution is partially circular: rules are generated from test predictions and then 'validated' by confidence computed on the same test predictions, so the reported 0.9+ confidences and the stability of the 14 rules are internal consistency checks rather than external evidence. This raises the circularity score to 4. Separately, the paper has serious non-circular correctness problems that should be weighed in the verdict: Table II reports ANSR-DT precision 0.80±0.05, recall 0.75±0.04, F1 0.77, ROC-AUC 0.85±0.03, while Fig. 6a's confusion matrix (343 TN, 0 FP, 153 TP, 3 FN) implies precision 1.00, recall ≈0.981, F1 ≈0.990, and Fig. 6b reports AP=1.00; no protocol is given that reconciles these values. The abstract also promises eight baselines and SKAB validation that do not appear in the body. These are completeness and consistency concerns, not circularity, and they make the headline performance claim unsupported as written.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The framework's reported performance depends on several hand-set coefficients and on the assumption that a self-generated synthetic dataset represents real industrial signals. The rule confidence formula and reward weights are introduced by the authors without external validation, and the single baseline comparison is not enough to support the 'eight baselines' claim in the abstract. No new physical entities are introduced.

free parameters (5)
  • class_weight_0 = 64.7
    Chosen to counteract the 5% event rate; it strongly shapes the precision/recall balance and the reported recall gain.
  • class_weight_1 = 0.5
    Minority-class weight in the loss function; together with class_weight_0 it drives the confusion matrix numbers.
  • confidence_threshold_tau = 0.85
    Rule incorporation threshold in Eq. 1; directly determines the 14-rule count and the stability claim.
  • reward weights alpha1, alpha2, alpha3 = 0.5, 0.3, 0.2
    Hand-set weights in Eq. 5 define what 'good' behavior means for the PPO agent and therefore shape the explained variance results.
  • Bayesian-optimized hyperparameters = learning rate 1e-3, dropout 0.3
    Tuned on validation data; reporting the resulting peak validation accuracy as the headline performance is a form of post hoc selection.
assumptions (4)
  • domain assumption The synthetic dataset with 5% injected events and correlations 0.3 to 0.5 represents realistic industrial operating conditions.
    All performance claims rest on this dataset; the KS/AD tests only check internal distributions, not fidelity to real industrial signals.
  • ad hoc to paper Rule confidence as defined in Eq. 4 is a valid measure of rule quality for extraction and conflict resolution.
    The formula is introduced by the authors without derivation or external benchmark, yet it determines which 14 rules enter the knowledge base.
  • domain assumption The single CNN-LSTM baseline is a sufficient comparison to support the 'competitive performance' and 'eight baselines' claims.
    Only one baseline appears in the body; the fairness and strength of the comparison are unverified.
  • ad hoc to paper The reward function weights (0.5, 0.3, 0.2) reflect the true operational priorities of industrial digital twins.
    The weights are hand-set with no sensitivity analysis or stakeholder validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ANSR-DT: A Neuro-Symbolic Framework for Adaptive and Explainable Digital Twins." pith.science (2026). https://pith.science/paper/BYWEOS7A

@misc{pith2026250108561,
  author       = {Pith},
  title        = {Pith review of: ANSR-DT: A Neuro-Symbolic Framework for Adaptive and Explainable Digital Twins},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BYWEOS7A}},
  note         = {Machine review of arXiv:2501.08561}
}
read the original abstract

Digital twins are increasingly used to monitor and optimize industrial systems, yet many existing frameworks remain difficult to interpret, slow to adapt, and limited in their ability to incorporate explicit domain knowledge. This paper presents ANSR-DT, an adaptive neuro-symbolic framework that unifies temporal anomaly detection, symbolic reasoning, and reinforcement-learning-based decision support within a single digital twin pipeline. ANSR-DT combines a CNN-LSTM model for multivariate pattern recognition with Prolog-based reasoning that converts learned signals into explicit rules, enabling transparent diagnoses and traceable decision paths. A PPO-based adaptation layer further refines operational responses under changing conditions while preserving interpretability. Experiments against eight baselines show that ANSR-DT delivers competitive predictive performance together with stable rule extraction, scalable symbolic reasoning, and actionable explanations. Additional validation on the Skoltech Anomaly Benchmark (SKAB) further indicates that the framework transfers beyond synthetic settings. These findings position ANSR-DT as a practical foundation for trustworthy, adaptive, and explainable industrial digital twins.

Figures

Figures reproduced from arXiv: 2501.08561 by the authors.

Figure 1
Figure 1. Overview of the ANSR-DT framework architecture. The framework consists of three main layers: (1) Physical Industrial Environment for sensor integration and human operator interaction, (2) Processing Layer implementing the neuro-symbolic reasoning engine with deep learning and symbolic components, and (3) Adaptation Layer incorporating reinforcement learning and dynamic rule updating mechanisms. Solid arrows indicate… view at source ↗
Figure 2
Figure 2. System operation sequence of the ANSR-DT framework. The diagram illustrates the interaction between components such as the Sensor Network, Data Manager, ML Module, Rule Engine, Digital Twin, and User Interface for real-time time-series pattern extraction and adaptive operations. digital components through a data preprocessing pipeline and feedback control loop. The data preprocessing pipeline systematically validate… view at source ↗
Figure 3
Figure 3. Multi-faceted visualization of synthetic sensor data: (a) Vibration measurements (mm/s) over operational hours showing operational thresholds, gradual events, and other key events; (b) Distribution of vibration sensor values distinguishing between nor￾mal operational data and key events; (c) Pressure readings (kPa) illustrating dynamic changes in system state during specific time steps. These visualizations demonstr… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: ANSR-DT framework architecture with three layers: Physical Environment for sensor integration, Processing Layer for [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Training and validation accuracy and loss trends over 20 epochs, showing convergence with some fluctuations. Validation accuracy reached a peak of 99.5% by epoch 15, with class weighting (0: 64.7, 1: 0.5) addressing imbalance in the dataset. rigorous comparison framewo…
Figure 6
Figure 6. Figure 6: Performance Metrics: (a) Confusion Matrix showing clas￾sification accuracy with 343 true negatives (100.0%) and 153 true positives (98.1%), with minimal false negatives (1.9%) and no false positives; (b) Precision-Recall Curve with an Average Precision (AP) score of 1.…
Figure 7
Figure 7. Figure 7 [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Focused ANSR-DT Knowledge Graph snapshot illustrating the diagnosis of a critical system event. The central State node (green) indicates “Critical” (C), directly linked to the poor performance Metrics (gray; E=0.07, P=7.2) and the high-confidence key event detected (sa…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 30 canonical work pages

  1. [1]

    Digital twin market - global forecast to 2028,

    MarketsandMarkets, “Digital twin market - global forecast to 2028,” MarketsandMarkets, Tech. Rep., 2023. [On- line]. Available: https://www.marketsandmarkets.com/Market-Reports/ digital-twin-market-225269522.html

    work page 2023

  2. [2]

    Digital twin in smart manufacturing,

    L. Li, B. Lei, and C. Mao, “Digital twin in smart manufacturing,” Journal of Industrial Information Integration, vol. 26, p. 100289, 2022

    work page 2022

  3. [3]

    V ohra,Digital twin technology: fundamentals and applications

    M. V ohra,Digital twin technology: fundamentals and applications. John Wiley & Sons, 2023

    work page 2023

  4. [4]

    Applications of digital twin across industries: A review,

    M. Singh, R. Srivastava, E. Fuenmayor, V . Kuts, Y . Qiao, N. Murray, and D. Devine, “Applications of digital twin across industries: A review,” Applied Sciences, vol. 12, no. 11, p. 5727, 2022

    work page 2022

  5. [5]

    Towards adaptive digital twins architecture,

    R. Ogunsakin, N. Mehandjiev, and C. A. Marin, “Towards adaptive digital twins architecture,”Computers in Industry, vol. 149, p. 103920, 2023

    work page 2023

  6. [6]

    Dynamic digital twins for situation awareness,

    E. Blasch, P. Schrader, G. Chen, S. Wei, Y . Chen, S. Khan, A. Aved, and A. Munir, “Dynamic digital twins for situation awareness,” inNAECON 2024-IEEE National Aerospace and Electronics Conference. IEEE, 2024, pp. 433–440

    work page 2024

  7. [7]

    Human digital twin in the context of industry 5.0,

    B. Wang, H. Zhou, X. Li, G. Yang, P. Zheng, C. Song, Y . Yuan, T. Wuest, H. Yang, and L. Wang, “Human digital twin in the context of industry 5.0,”Robotics and Computer-Integrated Manufacturing, vol. 85, p. 102626, 2024

    work page 2024

  8. [8]

    Methodology for enabling dynamic digital twins and virtual model evolution in industrial robotics-a predictive maintenance application,

    P. Aivaliotis, Z. Arkouli, K. Georgoulias, and S. Makris, “Methodology for enabling dynamic digital twins and virtual model evolution in industrial robotics-a predictive maintenance application,”International Journal of Computer Integrated Manufacturing, vol. 36, no. 7, pp. 947– 965, 2023

    work page 2023

Show all 33 references
  1. [9]

    Towards efficient neuro-symbolic ai: From workload characterization to hardware architecture,

    Z. Wan, C.-K. Liu, H. Yang, R. Raj, C. Li, H. You, Y . Fu, C. Wan, S. Li, Y . Kimet al., “Towards efficient neuro-symbolic ai: From workload characterization to hardware architecture,”IEEE Transactions on Circuits and Systems for Artificial Intelligence, 2024

    2024

  2. [10]

    Prox- imal policy optimization algorithms,

    J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, “Prox- imal policy optimization algorithms,”arXiv preprint arXiv:1707.06347, 2017

    2017 arXiv

  3. [11]

    Aˆ 3: Accelerating attention mechanisms in neural networks with approximation,

    T. J. Ham, S. J. Jung, S. Kim, Y . H. Oh, Y . Park, Y . Song, J.-H. Park, S. Lee, K. Park, J. W. Leeet al., “Aˆ 3: Accelerating attention mechanisms in neural networks with approximation,” in2020 IEEE International Symposium on High Performance Computer Architecture (HPCA). IE...

    2020

  4. [12]

    A re- view of explainable artificial intelligence in supply chain management using neurosymbolic approaches,

    E. E. Kosasih, E. Papadakis, G. Baryannis, and A. Brintrup, “A re- view of explainable artificial intelligence in supply chain management using neurosymbolic approaches,”International Journal of Production Research, vol. 62, no. 4, pp. 1510–1540, 2024

    2024

  5. [13]

    Neuro-symbolic explainable artificial intelligence twin for zero-touch ioe in wireless network,

    M. S. Munir, K. T. Kim, A. Adhikary, W. Saad, S. Shetty, S.-B. Park, and C. S. Hong, “Neuro-symbolic explainable artificial intelligence twin for zero-touch ioe in wireless network,”IEEE Internet of Things Journal, 2023

    2023

  6. [14]

    Reinforcement learning and autonomous driving: Comparison between dqn and ppo,

    D. Ma, “Reinforcement learning and autonomous driving: Comparison between dqn and ppo,” inAIP Conference Proceedings, vol. 3144, no. 1. AIP Publishing, 2024

    2024

  7. [15]

    Adaptive digital twins for energy-intensive industries and their local communities,

    T. G. Walmsley, P. Patros, W. Yu, B. R. Young, S. Burroughs, M. Ap- perley, J. K. Carson, I. A. Udugama, H. Aeowjaroenlap, M. J. Atkins et al., “Adaptive digital twins for energy-intensive industries and their local communities,”Digital Chemical Engineering, vol. 10, p. 100139, 2024

    2024

  8. [16]

    Systematic literature review on neuro-symbolic ai in knowledge graph construction for manufacturing,

    W. J. Schmidt, D. Rincon-Yanez, E. Kharlamov, and A. Paschke, “Systematic literature review on neuro-symbolic ai in knowledge graph construction for manufacturing,”Semantic Web Journal TBD, 2024

    2024

  9. [17]

    Neuro-symbolic ai at bosch: Data foundation, insights, and deployment

    B. Zhou, Z. Tan, Z. Zheng, D. Zhou, Y . He, Y . Zhu, M. Yahya, T.-K. Tran, D. Stepanova, M. H. Gad-Elrabet al., “Neuro-symbolic ai at bosch: Data foundation, insights, and deployment.” inISWC (Posters/Demos/Industry), 2022

    2022

  10. [18]

    Neurosymbolic ai: Combining deep learning and rule learning

    M. Jacob, L. Dierckx, and S. Nijssen, “Neurosymbolic ai: Combining deep learning and rule learning.”

  11. [19]

    A consistency method for digital twin model of human-robot collaboration,

    X. Ma, Q. Qi, J. Cheng, and F. Tao, “A consistency method for digital twin model of human-robot collaboration,”Journal of Manufacturing Systems, vol. 65, pp. 550–563, 2022

    2022

  12. [20]

    Accelerating proximal policy optimization learning using task prediction for solving games with delayed rewards,

    A. Ahmad, M. Kermanshah, K. Leahy, Z. Serlin, H. C. Siu, M. Mann, C.-I. Vasile, R. Tron, and C. Belta, “Accelerating proximal policy optimization learning using task prediction for solving games with delayed rewards,”arXiv preprint arXiv:2411.17861, 2024

    2024 arXiv

  13. [21]

    Developing a conceptual framework for the application of digital twin technologies to revamp building operation and maintenance processes,

    J. Zhao, H. Feng, Q. Chen, and B. G. de Soto, “Developing a conceptual framework for the application of digital twin technologies to revamp building operation and maintenance processes,”Journal of Building Engineering, vol. 49, p. 104028, 2022

    2022

  14. [22]

    A survey on verification and validation, testing and evaluations of neurosymbolic artificial intelligence,

    J. Renkhoff, K. Feng, M. Meier-Doernberg, A. Velasquez, and H. H. Song, “A survey on verification and validation, testing and evaluations of neurosymbolic artificial intelligence,”IEEE Transactions on Artificial Intelligence, 2024

    2024

  15. [23]

    A review of unit level digital twin applications in the manufacturing industry,

    T. B ¨ottjer, D. Tola, F. Kakavandi, C. R. Wewer, D. Ramanujan, C. Gomes, P. G. Larsen, and A. Iosifidis, “A review of unit level digital twin applications in the manufacturing industry,”CIRP Journal of Manufacturing Science and Technology, vol. 45, pp. 162–189, 2023

    2023

  16. [24]

    A practical approach to reconstruct high-quality landsat ndvi time-series data by gap filling and the savitzky–golay filter,

    Y . Chen, R. Cao, J. Chen, L. Liu, and B. Matsushita, “A practical approach to reconstruct high-quality landsat ndvi time-series data by gap filling and the savitzky–golay filter,”ISPRS Journal of Photogrammetry and Remote Sensing, vol. 180, pp. 174–190, 2021

    2021

  17. [25]

    Explainable ai for digital twins via neuro-symbolic rule extraction,

    S. B. Hakim, M. Adil, A. Velasquez, and H. H. Song, “Explainable ai for digital twins via neuro-symbolic rule extraction,” in2025 IEEE 22nd International Conference on Ubiquitous Intelligence and Computing (UIC). IEEE, 2025

    2025

  18. [26]

    Unsupervised learning of neuro- symbolic rules for generalizable context-aware planning in object ar- rangement tasks,

    S. Sharma, S. Tuli, and R. Paul, “Unsupervised learning of neuro- symbolic rules for generalizable context-aware planning in object ar- rangement tasks,” in2024 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2024, pp. 12 865–12 872

    2024

  19. [27]

    Problog 2: A probabilistic logic programming toolbox,

    M.-K. Leuven, “Problog 2: A probabilistic logic programming toolbox,” 2023, available at https://github.com/ML-KULeuven/problog. [Online]. Available: https://github.com/ML-KULeuven/problog

    2023

  20. [28]

    A deep learning-enhanced digital twin framework for improving safety and reliability in human–robot collaborative manufacturing,

    S. Wang, J. Zhang, P. Wang, J. Law, R. Calinescu, and L. Mihaylova, “A deep learning-enhanced digital twin framework for improving safety and reliability in human–robot collaborative manufacturing,”Robotics and computer-integrated manufacturing, vol. 85, p. 102608, 2024

    2024

  21. [29]

    Anderson- 14 darling and watson tests for the geometric distribution with estimated probability of success,

    H. F. Coronel-Brizio, A. R. Hern ´andez-Montoya, M. E. Rodr ´ıguez- Achach, H. Tapia-McClung, and J. E. Trinidad-Segovia, “Anderson- 14 darling and watson tests for the geometric distribution with estimated probability of success,”PloS one, vol. 19, no. 12, p. e0315855, 2024

    2024

  22. [30]

    Automatic tuning of hyperparame- ters using bayesian optimization,

    A. H. Victoria and G. Maragatham, “Automatic tuning of hyperparame- ters using bayesian optimization,”Evolving Systems, vol. 12, no. 1, pp. 217–223, 2021

    2021

  23. [31]

    Evaluating the robustness of neural networks: An extreme value theory approach,

    T.-W. Weng, H. Zhang, P.-Y . Chen, J. Yi, D. Su, Y . Gao, C.-J. Hsieh, and L. Daniel, “Evaluating the robustness of neural networks: An extreme value theory approach,”arXiv preprint arXiv:1801.10578, 2018

    2018 arXiv

  24. [32]

    Stable-baselines3: Reliable reinforcement learning implementations,

    A. Raffin, A. Hill, A. Gleave, A. Kanervisto, M. Ernestus, and N. Dormann, “Stable-baselines3: Reliable reinforcement learning implementations,”Journal of Machine Learning Research, vol. 22, no. 268, pp. 1–8, 2021. [Online]. Available: http://jmlr.org/papers/v22/ 20-1364.html

    2021

  25. [33]

    Digital-twin-based real- time optimization for a fractional order controller for industrial robots,

    X. Liu, H. Gan, Y . Luo, Y . Chen, and L. Gao, “Digital-twin-based real- time optimization for a fractional order controller for industrial robots,” Fractal and Fractional, vol. 7, no. 2, p. 167, 2023

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.