Pith. sign in

REVIEW 4 major objections 6 minor 28 references

Multimodal Anomaly Detection with a Mixture-of-Experts

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Confidence-based fusion of force and vision detectors cuts robot anomaly delay by 60%.

desk verdict A sensible two-expert fusion for robot anomaly detection, but Eq. 5 inverts GMR confidence and the printed fusion rule contradicts the reported results. read the letter →

arxiv 2506.19077 v1 pith:KUPTJVFN submitted 2025-06-23 cs.RO

Christoph Willibald , Daniel Sliwowski , Dongheui Lee This is my paper · ORCID
classification cs.RO
keywords anomalydetectionmixture-of-expertsrobotmanipulationGaussianmixtureregressionvision-languagemodelconfidence-basedfusionexecutionmonitoringlate
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

Robotic anomaly detection usually comes in two incompatible flavours: statistical detectors that flag deviations in forces and motions, and vision-based detectors that spot wrong scene states. This paper tries to establish that combining one of each into a two-expert mixture, with a winner-takes-all rule that trusts whichever expert reports higher confidence, catches both kinds of failure faster than either detector alone. In a box-grasping task the fused system cut detection delay by more than 60% relative to both individual detectors while matching their frame-wise accuracy, and in a juice-pouring task it improved every reported frame-wise score. The reason to care is practical: robots operating around people need one monitoring system that responds to both internal faults and external disturbances without retraining on combined data.

What carries the argument

The load-bearing mechanism is the confidence-based winner-takes-all fusion rule, Eq. (8): $P = P_{\mathrm{GMR}}$ if $C_{\mathrm{GMR}} > C_{\mathrm{VLM}}$, otherwise $P = P_{\mathrm{VLM}}$. For the GMR expert, confidence $C_{\mathrm{GMR}}$ is a sigmoid-scaled version of the normalized Mahalanobis distance $\varepsilon_t = D_{M,t}/D^K_{M,\max}$, where $D_{M,t}$ is the distance between measured and regressed output features and $D^K_{M,\max}$ is the largest such distance seen in training for the selected mixture component; the scale factor $\alpha$ is chosen by hand. For the VLM expert, confidence $C_{\mathrm{VLM}}$ is the model's probability assigned to the expected action stage (divided by the number of stages in the expected set) when the prediction matches, or the summed probability over anomalous classes when it does not. The phase variable $s_t$ of a Dynamical Movement Primitive defines the expected stage $g(s_t)$, which lets the system monitor continuous motion policies instead of requiring segmented primitive motions. The confidence scores are what make the fusion dynamic: when a force deviation is imminent but not yet visually evident, GMR's confidence rises first, while semantic scene errors such as a spill raise the VLM's confidence, so the rule selects the expert that can see the fault earliest.

What would settle it

Record, for a held-out set of executions, the confidence pair $(C_{\mathrm{GMR}}, C_{\mathrm{VLM}})$ and the correctness of each expert's prediction at every frame, then check whether the expert with higher confidence is correct more often than the lower-confidence expert. If there is a substantial region where the higher-confidence expert is systematically wrong while the lower-confidence one is right, the fusion rule fails; the same test on the two reported tasks would already give evidence either way.

Watch

Extended reading notes

Core claim

The paper's central claim is that a late-fusion mixture-of-experts can outperform its two constituents on robotic manipulation anomaly detection. The first expert is a Gaussian-mixture-regression detector: it learns expected contact forces and end-effector poses from successful demonstrations, measures the Mahalanobis distance between predicted and measured values, and flags an anomaly when that distance exceeds a per-modality threshold. The second expert is a vision-language model that is given the current camera frame and a natural-language task description, predicts whether the action's precondition, effect, or neither is satisfied, and compares that prediction against the stage expected from the motion phase. The fusion rule is simple: each expert computes a confidence score, and the expert with the higher confidence supplies the final anomaly decision. The authors report that this rule selects the more appropriate detector in the scenarios they test, giving a 60% detection-delay reduction in the industrial box-grasping task and better or equal frame-wise accuracy, precision, recall, and F1 scores in both tasks.

Load-bearing premise

The load-bearing premise is that the two confidence scores are commensurable, so the higher number really means 'this expert is more likely to be right'; the paper selects the sigmoid scale and the probability sums without a calibration step, and if the scales are off, the winner-takes-all rule can systematically pick the weaker detector.

Editorial extensions

If this is right

  • In contact-rich industrial skills, anomalies linked to force or trajectory deviations can be flagged up to 60% earlier than with a single detector, giving the robot or a supervisor more time to intervene.
  • Anomalies that leave no force trace, such as liquid dripping from a bottle, become detectable in the same fused system because the vision-language expert carries those cases.
  • Force-only false positives can be overruled by high-confidence visual predictions, and visual false alarms can be overruled by high-confidence force predictions, as shown in the pouring-task plots.
  • The two experts can be trained on separate demonstration subsets collected by different methods (kinesthetic teaching for forces, teleoperation for visuals), reducing the need for one dataset that satisfies both modalities.
  • Adding more modalities such as audio or depth, or additional expert detectors, is a direct extension of the same late-fusion structure, as the paper itself suggests.

Reading between the lines

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

  • I infer that the winner-takes-all rule is only as good as the comparability of the two confidence scores; since the GMR score uses a hand-chosen sigmoid scale and the VLM score a softmax-derived sum, a systematic scale mismatch could make the rule consistently trust the wrong expert, and the paper does not test this directly.
  • I infer that the expected-stage function $g(s_t)$ is specified by hand for each skill, so the claimed benefits depend on the effort and correctness of that manual specification; automating $g$ from demonstrations would be a natural stress test.
  • A testable extension would be to measure how often the chosen expert's confidence ranking matches its actual prediction correctness across frames; if the two disagree often, a learned gating network or calibrated probabilities would likely outperform the raw winner-takes-all choice.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. This paper presents a multimodal anomaly detection framework for robotic manipulation that fuses two complementary detectors: a Gaussian-mixture-regression (GMR) model that monitors proprioceptive signals (end-effector pose and contact forces) and a vision-language model (ConditionNET) that classifies visual frames against expected action preconditions and effects. The two experts are combined by a winner-takes-all rule that selects the prediction of the expert with the higher confidence score. The framework is evaluated on a box-grasping/locking task and a juice-pouring task, with metrics including frame-wise accuracy, precision, recall, F1, F1@50% overlap, and detection delay. The authors report a 60% reduction in detection delay in the box-grasping task and improved or comparable frame-wise scores relative to the individual detectors.

Significance. The core idea of combining a low-level statistical detector with a high-level semantic VLM detector is timely and practically motivated: robot-driven and environment-driven anomalies have complementary signatures, and the paper provides one of the first explicit confidence-based fusion mechanisms for these two families. The evaluation covers two physically different tasks and multiple anomaly types, and the inclusion of detection delay is a useful practical metric. The method is conceptually simple and, once the confidence definition is corrected, could be straightforwardly adapted by other groups. However, the central confidence equation appears to be inverted, the fusion-rule comparability assumption is unvalidated, and the experiments lack statistical grounding and a successful-execution test set for the pouring task; these issues currently prevent acceptance.

major comments (4)
  1. [III.B, Eq. (5)] The GMR confidence score defined in Eq. (5) is inverted relative to its use in Eq. (8). For ε_t > 1, C_GMR = σ(−α(ε_t − 1)) decreases monotonically from 0.5 at the decision boundary to 0 as ε_t grows; for ε_t < 1 it also decreases as ε_t moves away from 1. Confidence is therefore highest exactly at the anomaly threshold and lowest for the most clear-cut anomalies. Under the winner-takes-all rule in Eq. (8), the MoE would suppress the GMR expert when its Mahalanobis deviation is largest, which contradicts the qualitative description in Section IV.C ('higher prediction confidence score of the GMR-based detector') and makes the reported 60% delay reduction unreproducible from the printed equations. Please correct the signs (likely C_GMR = σ(α(ε_t − 1)) for ε_t > 1 and σ(−α(ε_t − 1)) for ε_t ≤ 1) and re-verify all experiments and figures affected by this definition.
  2. [III.D, Eqs. (5)-(8)] The fusion rule assumes that the confidence outputs of the two experts are directly comparable, but the manuscript provides no calibration or justification. GMR confidence is a sigmoid-transformed Mahalanobis ratio with a hand-chosen α, whereas VLM confidence is a softmax probability, in the no-anomaly case scaled by 1/|g(s_t)| and in the anomaly case summed over anomalous classes. These scores have different ranges and semantics, so the max operator in Eq. (8) may systematically select the less reliable expert. Please provide an analysis of the confidence scales (e.g., calibration curves, a swept threshold, or a per-task comparison of the fusion output vs. either expert alone) to support the claim that confidence-based selection is valid.
  3. [IV.A, IV.B, Table I (Pouring)] The pouring evaluation is performed exclusively on 27 failed autonomous executions; no successful pouring executions are included in the test set. Consequently, false alarms during normal, complete task executions are never measured, and the precision and F1 values reported in Table I are computed only over negative frames embedded in known-failure runs. This setup is biased toward the detection setting and does not support the claim of improved frame-wise performance in the pouring task. Add successful test executions (or an explicit statement about the intended deployment regime) and report metrics that include them.
  4. [IV.B, Table I] All quantitative results are reported as single aggregate values without variance, confidence intervals, or significance tests. The test sets are small (82 skill executions for box grasping, 27 for pouring), and the headline claims (60% delay reduction, F1 improvements) are based on differences that could easily arise from a few outlier executions. Please report per-execution statistics or bootstrap confidence intervals and, where appropriate, perform a paired significance test between methods.
minor comments (6)
  1. [Abstract / Table I] The abstract states that the method reduces detection delay by 60% while improving frame-wise detection performance compared to individual detectors; in the pouring task MoE's delay (-0.3 s) is actually slightly worse than GMR (-0.4 s), and in the box task MoE accuracy (88.1) is lower than GMR (88.8). Please qualify these claims to the specific task and metric.
  2. [III.D, Eq. (8)] Eq. (8) uses the notation PGMM/CGMM while the rest of the paper uses PGMR/CGMR; please unify the notation.
  3. [III.B] The hyperparameter α, the number of mixture components K, and the sliding-window length (eight time steps) are introduced without a selection criterion or sensitivity analysis; a short paragraph or ablation would help the reader assess robustness.
  4. [III.C / IV] The expected-stage function g(s_t) is manually specified for each skill; the manuscript should state how sensitive the VLM-based predictions are to this design choice and ideally provide the mappings used in the experiments.
  5. [IV.B] The negative detection delay values in the pouring task (MoE -0.3, GMR -0.4) should be explicitly defined (e.g., detection before the ground-truth onset) and it should be stated whether such early alarms are always desirable.
  6. [Fig. 4/5 captions] There is a minor grammar/capitalization issue in the captions ('If the Mahalanobis distance exceeds the threshold, The GMR-based detector triggers an anomaly'), and the red and green curves may be hard to distinguish in grayscale.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the winner-takes-all fusion is an empirical selection rule; the claimed delay reduction is not forced by the defining equations, and the self-citations supply components that are also independently evaluated baselines.

full rationale

The derivation chain is: (i) GMR learns a GMM from successful demonstrations and computes a Mahalanobis ratio epsilon_t = D_M,t / D^k_{M,max} (Eq. 3), thresholding at 1 (Eq. 4) and assigning a sigmoidal confidence (Eq. 5); (ii) the VLM outputs a class distribution, and the expected class is specified by a manually chosen phase function g(s_t), giving P_VLM and C_VLM (Eqs. 6-7); (iii) Eq. 8 selects the expert with larger confidence. The final anomaly label is therefore defined as one of the two experts' labels; it is not defined in terms of the evaluation metrics (frame-wise F1, detection delay), and no parameter of the fusion rule is fit to the test executions used to report Table I. The GMR threshold D^k_{M,max} is fit to training maxima, but that is a standard unsupervised threshold and is not renamed as a prediction. The paper builds on the authors' prior detectors [2], [8], [27], but those are also the baselines against which MoE is measured, and the qualitative mechanism (e.g., GMR detecting force deviations early, VLM detecting visual-only spills) is described from the data rather than imported as a theorem. The apparent inversion in Eq. 5, where C_GMR decreases as epsilon_t grows beyond 1, is a consistency or correctness concern that would undermine the qualitative claim in Fig. 4a, but it is not a circularity: the confidence formula does not incorporate the benchmark results. No load-bearing step reduces to its own input, so the appropriate finding is no significant circularity.

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

The central claim rests on four assumptions and four free parameters. The most significant are the manual phase-to-state mapping g(s_t), the confidence comparability assumption, and thresholds fit to training data maxima. No new physical entities are postulated.

free parameters (4)
  • alpha (GMR confidence scaling) = not reported
    Hyperparameter in Eq. (5) controlling the slope of the sigmoid confidence function; chosen by hand per task with no sensitivity analysis.
  • GMM component count K = 2 (box), 10 (pouring)
    Number of Gaussian components for the GMM/GMR model, chosen manually per skill.
  • Dk_M,max per-component Mahalanobis thresholds = not listed numerically; set to maximum training Mahalanobis distance per component
    Anomaly thresholds in Eq. (4) are the maximum observed Mahalanobis distance on the (small) training set for each mixture component, so they are fitted values.
  • Sliding-window length for majority filtering = 8 time steps
    Raw predictions are smoothed by majority vote over 8 steps; window size chosen without ablation.
assumptions (4)
  • domain assumption The DMP phase variable s is a sufficient task-progress measure, and the manually specified mapping g(s_t) correctly describes the expected action state at every phase.
    Introduced in Section III.C: 'we define the expected model predictions as a function of the phase variable, g(s_t), and manually specify it for each skill.' If g is wrong or phase does not track semantic progress, the VLM expert mislabels anomalies.
  • domain assumption The GMM/GMR conditional distribution P(xi_O | xi_I), conditioned on end-effector pose, captures all relevant expected force/motion dynamics, so Mahalanobis distance (Eq. 3) is a valid anomaly score.
    Section III.B: the output features are compared against the conditional mean and covariance. This assumes pose-to-force mapping is unimodal within each Gaussian component and that anomalies manifest as deviations in the chosen output features.
  • domain assumption ConditionNET (VLM) trained on the (Im)PerfectPour dataset and on 107 successful/71 unsuccessful box examples transfers to the evaluation camera views and robot embodiments.
    The VLM is trained on demonstrations collected via teleoperation in prior work [8]; the paper applies it to new autonomous executions without reporting domain-shift analysis.
  • ad hoc to paper Confidence scores from the sigmoid-scaled Mahalanobis ratio and the VLM softmax probabilities are commensurable, so a simple max-confidence selection (Eq. 8) picks the better detector.
    No calibration is performed; the GMR confidence uses a hand-chosen alpha and the VLM confidence uses raw probabilities or sums of probabilities. The fusion rule's correctness depends on this comparability.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multimodal Anomaly Detection with a Mixture-of-Experts." pith.science (2026). https://pith.science/paper/KUPTJVFN

@misc{pith2026250619077,
  author       = {Pith},
  title        = {Pith review of: Multimodal Anomaly Detection with a Mixture-of-Experts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KUPTJVFN}},
  note         = {Machine review of arXiv:2506.19077}
}
read the original abstract

With a growing number of robots being deployed across diverse applications, robust multimodal anomaly detection becomes increasingly important. In robotic manipulation, failures typically arise from (1) robot-driven anomalies due to an insufficient task model or hardware limitations, and (2) environment-driven anomalies caused by dynamic environmental changes or external interferences. Conventional anomaly detection methods focus either on the first by low-level statistical modeling of proprioceptive signals or the second by deep learning-based visual environment observation, each with different computational and training data requirements. To effectively capture anomalies from both sources, we propose a mixture-of-experts framework that integrates the complementary detection mechanisms with a visual-language model for environment monitoring and a Gaussian-mixture regression-based detector for tracking deviations in interaction forces and robot motions. We introduce a confidence-based fusion mechanism that dynamically selects the most reliable detector for each situation. We evaluate our approach on both household and industrial tasks using two robotic systems, demonstrating a 60% reduction in detection delay while improving frame-wise anomaly detection performance compared to individual detectors.

Figures

Figures reproduced from arXiv: 2506.19077 by the authors.

Figure 1
Figure 1. Our Mixture of Experts (MoE) framework that [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. System Architecture. After training the VLM and GMR experts, we developed a mixture-of-experts anomaly detection pipeline that combines their strengths. For the GMR expert, we use measured input features to regress the predicted feature value µˆ O t and covariance Σˆ O t . We compute the maximum Mahalanobis distance for the mixture component associated with the observed input features and the Mahalanobis distance be… view at source ↗
Figure 3
Figure 3. The different phases and challenges of the box [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Anomaly prediction results for different anomaly scenarios in the pouring task. The first two rows show the computed [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Anomaly prediction results for different anomaly [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 19 canonical work pages

  1. [1]

    Anomaly detection: A survey,

    V . Chandola, A. Banerjee, and V . Kumar, “Anomaly detection: A survey,” ACM Comput. Surv. , vol. 41, no. 3, July 2009. [Online]. Available: https://doi.org/10.1145/1541880.1541882

    arXiv 2009

  2. [2]

    Collaborative programming of conditional robot tasks,

    C. Willibald, T. Eiband, and D. Lee, “Collaborative programming of conditional robot tasks,” in 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2020, pp. 5402– 5409

    work page 2020

  3. [3]

    Collabo- rative programming of robotic task decisions and recovery behaviors,

    T. Eiband, C. Willibald, I. Tannert, B. Weber, and D. Lee, “Collabo- rative programming of robotic task decisions and recovery behaviors,” Autonomous Robots, vol. 47, no. 2, pp. 229–247, 2023

    work page 2023

  4. [4]

    Intuitive programming of con- ditional tasks by demonstration of multiple solutions,

    T. Eiband, M. Saveriano, and D. Lee, “Intuitive programming of con- ditional tasks by demonstration of multiple solutions,” IEEE Robotics and Automation Letters , vol. 4, no. 4, pp. 4483–4490, 2019

    work page 2019

  5. [5]

    Anomaly detection for insertion tasks in robotic assembly using gaussian process models,

    D. Romeres, D. K. Jha, W. Yerazunis, D. Nikovski, and H. A. Dau, “Anomaly detection for insertion tasks in robotic assembly using gaussian process models,” in 2019 18th European Control Conference (ECC). IEEE, 2019, pp. 1017–1022

    work page 2019

  6. [6]

    Multimodal anomaly detection for assistive robots,

    D. Park, H. Kim, and C. C. Kemp, “Multimodal anomaly detection for assistive robots,” Autonomous Robots, vol. 43(3), pp. 611–629, 2019

    work page 2019

  7. [7]

    Hmms for anomaly detection in autonomous robots,

    D. Azzalini, A. Castellini, M. Luperto, A. Farinelli, and F. Amigoni, “Hmms for anomaly detection in autonomous robots,” in Int. Conf. on Autonomous Agents and MultiAgent Systems , 2020, pp. 105–113

    2020

  8. [8]

    Conditionnet: Learning preconditions and effects for execution monitoring,

    D. Sliwowski and D. Lee, “Conditionnet: Learning preconditions and effects for execution monitoring,” IEEE Robotics and Automation Letters, 2024

    work page 2024

Show all 28 references
  1. [9]

    A multimodal anomaly detector for robot-assisted feeding using an lstm-based variational autoen- coder,

    D. Park, Y . Hoshi, and C. C. Kemp, “A multimodal anomaly detector for robot-assisted feeding using an lstm-based variational autoen- coder,” IEEE Robotics and Automation Letters, vol. 3, no. 3, pp. 1544– 1551, 2018

    2018

  2. [10]

    Multimodal anomaly detection based on deep auto-encoder for object slip perception of mobile manipulation robots,

    Y . Yoo, C.-Y . Lee, and B.-T. Zhang, “Multimodal anomaly detection based on deep auto-encoder for object slip perception of mobile manipulation robots,” in 2021 IEEE International Conference on Robotics and Automation (ICRA) , 2021, pp. 11 443–11 449

    2021

  3. [11]

    Multimodal detection and classification of robot manipulation failures,

    A. Inceoglu, E. E. Aksoy, and S. Sariel, “Multimodal detection and classification of robot manipulation failures,” IEEE Robotics and Automation Letters, vol. 9, no. 2, pp. 1396–1403, 2024

    2024

  4. [12]

    Clue-ai: A convolutional three-stream anomaly identification framework for robot manipulation,

    D. Altan and S. Sariel, “Clue-ai: A convolutional three-stream anomaly identification framework for robot manipulation,” IEEE Access, vol. 11, pp. 48 347–48 357, 2022. [Online]. Available: https://api.semanticscholar.org/CorpusID:247476170

    2022

  5. [13]

    Robust, deep and in- ductive anomaly detection,

    R. Chalapathy, A. K. Menon, and S. Chawla, “Robust, deep and in- ductive anomaly detection,” in Joint European Conference on Machine Learning and Knowledge Discovery in Databases . Springer, 2017, pp. 36–51

    2017

  6. [14]

    Vision-language models as success detectors,

    Y . Du, K. Konyushkova, M. Denil, A. Raju, J. Landon, F. Hill, N. de Freitas, and S. Cabi, “Vision-language models as success detectors,” in Proceedings of The 2nd Conference on Lifelong Learning Agents , ser. Proceedings of Machine Learning Research, S. Chandar, R. Pascanu, H...

  7. [15]

    Fino-net: A deep multimodal sensor fusion framework for manipulation failure detection,

    A. Inceoglu, E. E. Aksoy, A. Cihan Ak, and S. Sariel, “Fino-net: A deep multimodal sensor fusion framework for manipulation failure detection,” in 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , 2021, pp. 6841–6847

    2021

  8. [16]

    Grounding classical task planners via vision-language models,

    X. Zhang, Y . Ding, S. Amiri, H. Yang, A. Kaminski, C. Esselink, and S. Zhang, “Grounding classical task planners via vision-language models,” in arXiv preprint arXiv:2304.08587 , 04 2023

    2023 arXiv

  9. [17]

    Real-Time Anomaly Detection and Reactive Planning with Large Language Models,

    R. Sinha, A. Elhafsi, C. Agia, M. Foutter, E. Schmerling, and M. Pavone, “Real-Time Anomaly Detection and Reactive Planning with Large Language Models,” in Proceedings of Robotics: Science and Systems, Delft, Netherlands, July 2024

    2024

  10. [18]

    Confidence-based policy learning from demonstration using gaussian mixture models,

    S. Chernova and M. Veloso, “Confidence-based policy learning from demonstration using gaussian mixture models,” in Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems, 2007, pp. 1–8

    2007

  11. [19]

    Unpacking failure modes of generative policies: Runtime monitoring of consistency and progress,

    C. Agia, R. Sinha, J. Yang, Z.-a. Cao, R. Antonova, M. Pavone, and J. Bohg, “Unpacking failure modes of generative policies: Runtime monitoring of consistency and progress,” arXiv preprint arXiv:2410.04640, 2024

    2024 arXiv

  12. [20]

    Autoencoder-based network anomaly detection,

    Z. Chen, C. K. Yeo, B. S. Lee, and C. T. Lau, “Autoencoder-based network anomaly detection,” in 2018 Wireless Telecommunications Symposium (WTS), 2018, pp. 1–5

    2018

  13. [21]

    Mad-gan: Multi- variate anomaly detection for time series data with generative adversar- ial networks,

    D. Li, D. Chen, B. Jin, L. Shi, J. Goh, and S.-K. Ng, “Mad-gan: Multi- variate anomaly detection for time series data with generative adversar- ial networks,” in Artificial Neural Networks and Machine Learning – ICANN 2019: Text and Time Series, I. V . Tetko, V . K˚urkov´a, P....

    2019

  14. [22]

    Pddl - the planning domain definition language,

    G. M. et al., “Pddl - the planning domain definition language,” Techincal Report, 08 1998

    1998

  15. [23]

    Incremental kinesthetic teaching of motion primitives using the motion refinement tube,

    D. Lee and C. Ott, “Incremental kinesthetic teaching of motion primitives using the motion refinement tube,” Autonomous Robots , vol. 31, no. 2, pp. 115–131, Oct 2011. [Online]. Available: https://doi.org/10.1007/s10514-011-9234-3

    2011 doi

  16. [24]

    Reassemble: A multimodal dataset for contact-rich robotic assembly and disassembly,

    D. Sliwowski, S. Jadav, S. Stanovcic, J. Orbik, J. Heidersberger, and D. Lee, “Reassemble: A multimodal dataset for contact-rich robotic assembly and disassembly,” arXiv preprint arXiv:2502.05086 , 2025

    2025 arXiv

  17. [25]

    Invariant description of rigid body motion trajectories,

    J. De Schutter, “Invariant description of rigid body motion trajectories,” Journal of Mechanisms and Robotics , vol. 2, no. 1, p. 011004, 11 2009. [Online]. Available: https://doi.org/10.1115/1.4000524

    2009 doi

  18. [26]

    Dynamic movement primitives in robotics: A tutorial survey,

    M. Saveriano, F. J. Abu-Dakka, A. Kramberger, and L. Peternel, “Dynamic movement primitives in robotics: A tutorial survey,” The International Journal of Robotics Research , vol. 42, no. 13, pp. 1133–1184, 2023. [Online]. Available: https://doi.org/10.1177/02783649231201196

    2023 doi

  19. [27]

    Multi-level task learning based on intention and constraint inference for autonomous robotic manipulation,

    C. Willibald and D. Lee, “Multi-level task learning based on intention and constraint inference for autonomous robotic manipulation,” 2022, pp. 7688–7695

    2022

  20. [232]

    PMLR, 22–25 Aug 2023, pp. 120–136. [Online]. Available: https://proceedings.mlr.press/v232/du23b.html

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.