Pith. sign in

REVIEW 2 major objections 5 minor 94 references

Behavioral uncertainty can rebalance rare unsafe labels for CPS safety monitors without synthesizing data.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-13 18:04 UTC pith:5A4E6ZQV

load-bearing objection Solid empirical methods paper: uncertainty-driven LNR beats fusion and standard rebalancers by a large margin on one hard UAV safety task; the flip premise is the real soft spot, not a fatal flaw. the 2 major comments →

arxiv 2603.25670 v4 pith:5A4E6ZQV submitted 2026-03-26 cs.LG cs.SE

Uncertainty-Guided Label Rebalancing for CPS Safety Monitoring

John Ayotunde , Qinghua Xu , Guancheng Wang , Lionel C. Briand This is my paper
classification cs.LG cs.SE
keywords Cyber-Physical SystemsSafety MonitoringLabel RebalancingBehavioral UncertaintyClass ImbalanceUAV TelemetryTime-Series Classification
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Safety predictors for cyber-physical systems fail when unsafe events are extremely rare, and standard rebalancing either invents unrealistic time-series samples or overfits the minority class. This paper argues that behavioral uncertainty—erratic control or rapid heading changes—correlates with safety risk and can be used to fix the imbalance. U-Balance first scores each telemetry window for uncertainty from simple distributional kinematic statistics via a gated MLP, then stochastically flips the labels of unusually uncertain safe windows to unsafe. A standard safety predictor trained on the resulting set reaches 0.806 F1 on a large UAV benchmark with a 46:1 imbalance, beating strong baselines by at least 14 percentage points. The claim matters because it turns an often-ignored operational signal into a practical lever for rare-event safety monitoring without fabricating data.

Core claim

On a large UAV safety-monitoring benchmark with extreme class imbalance, using predicted behavioral uncertainty to probabilistically relabel high-uncertainty safe windows as unsafe (uLNR) yields a safety predictor with F1 0.806, substantially better than early or late fusion of the same uncertainty signal and better than fourteen standard rebalancing methods and fourteen model baselines.

What carries the argument

Uncertainty-guided label rebalancing (uLNR): convert GatedMLP uncertainty scores of safe windows into z-scores, map them through a shifted tanh into flip probabilities, and stochastically relabel only those windows, thereby enriching the minority class with boundary samples without synthesizing new telemetry.

Load-bearing premise

Safe-labeled windows that show unusually high behavioral uncertainty are informative boundary cases that should be treated as unsafe for training, even though the ground-truth safety rule still marks them safe and the uncertainty-safety correlation is only moderate.

What would settle it

On the same UAV windows, if a held-out evaluation showed that the windows flipped by uLNR at the chosen threshold are predominantly true-safe (not near-miss or future-unsafe) and that removing those flips collapses F1 back to the plain or fusion baselines, the central claim would fail.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper proposes U-Balance for CPS safety monitoring under extreme class imbalance. It trains a GatedMLP uncertainty predictor on distributional kinematic features of telemetry windows, then applies uncertainty-guided label rebalancing (uLNR) that stochastically flips high-uncertainty safe labels to unsafe (Eqs. 7–8, §3.2), and finally trains a Bi-LSTM safety predictor on the rebalanced set. On the Khatiri et al. UAV window-level benchmark (≈46:1 safe:unsafe), the authors report a moderate point-biserial correlation between uncertainty and safety (r_pb = 0.444), show that uLNR substantially outperforms early/late fusion of uncertainty (Table 1), and claim F1 = 0.806 ± 0.031, a 14.3 pp gain over the strongest baseline (TimeMoE), with large Mann-Whitney effect sizes, competitive latency, and ablations of GatedMLP, distributional features, τ, and 14 rebalancers.

Significance. If the central premise holds—that high-uncertainty safe windows are informative boundary samples rather than pure noise—the work offers a practical, non-synthetic rebalancing route for rare-event CPS safety monitoring and a concrete alternative to fusion-based use of uncertainty. Strengths include a large multi-seed evaluation (30 runs), proper non-parametric testing with Â12, honest reporting of the precision–recall trade-off with τ (Table 4), a broad baseline suite (14 methods plus 14 rebalancers), and explicit comparison of uLNR against early/late fusion. The adaptation of LNR to time-series CPS data via a GatedMLP on distributional kinematics is a clear methodological contribution within the paper’s scope.

major comments (2)
  1. §3.2 / Eqs. 7–8 and §5.1: The load-bearing premise is that safe-labeled windows with high z-scored uncertainty are informative decision-boundary samples worth flipping to unsafe. The paper itself reports only moderate correlation (r_pb = 0.444 ± 0.014) and substantial overlap (many safe windows near maneuvers remain safe under the 1.5 m rule). At the chosen τ = 3.0, 1 140 training windows (~1.7 %) are flipped, raising the unsafe ratio only to 2.9 % (Table 4). Without a direct check of how many flipped windows are true-safe under the ground-truth distance rule (or of performance when those flips are replaced by random flips of the same cardinality / same z-score ranks), it remains possible that uLNR injects structured label noise that happens to improve recall on this simulation distribution. A short diagnostic—fraction of flipped windows that violate the 1.5 m rule, or a random-flip cont
  2. §4.2 / External validity: All results rest on a single Surrealist-generated PX4 UAV dataset with rule-based uncertainty and safety labels. The threats section notes that uncertainty can be auto-annotated elsewhere, but no second CPS domain, real-flight telemetry, or alternative safety definition is evaluated. Given that the F1 gain is large and the mechanism depends on the uncertainty–safety relationship, at least a limited transfer experiment (different windowing, different obstacle threshold, or another public CPS safety set with analogous uncertainty rules) is needed before the method can be recommended as a general CPS rebalancing technique.
minor comments (5)
  1. §3.1 / Eq. (1): The distributional feature vector is written as d_t ∈ R^16, but the four channels × four statistics are listed without an explicit concatenation operator; a one-line clarification would help.
  2. Figure 3 caption and body: “Proprocessing” is misspelled; also “U-Balanc3” appears as a stray heading fragment before §3.1.
  3. Table 2: Superialist is flight-level unsupervised reconstruction; a short note that its window-level comparison is therefore imperfect would avoid over-interpreting the large ΔF1.
  4. §4.5 / Implementation: Hyperparameter grids are deferred to a repository; for archival completeness, the final chosen values for all baselines should appear in an appendix or table.
  5. References and ACM template still contain placeholder conference metadata (Conference acronym ’XX, Woodstock, NY, 2018 dates).

Circularity Check

0 steps flagged

No significant circularity: uncertainty and safety labels are independently rule-derived, uLNR only rewrites training labels, and F1 is measured on held-out original safety labels.

full rationale

U-Balance is a standard supervised pipeline, not a first-principles derivation. Uncertainty labels (behavioral inconsistency rules from Khatiri et al.) and safety labels (1.5 m obstacle distance) are distinct and externally defined; the paper measures their moderate correlation (r_pb = 0.444) rather than assuming it into the metric. The GatedMLP is trained on uncertainty labels only; uLNR (Eqs. 7–8) uses its scores solely to set flip probabilities on majority-class training labels; the Bi-LSTM safety predictor is then trained on the rebalanced set and evaluated for precision/recall/F1 exclusively against the original held-out safety labels. No quantity is fitted and then re-presented as a prediction of a closely related quantity, no uniqueness theorem or ansatz is imported via self-citation, and no definitional identity equates the reported F1 to the rebalancing inputs. Self-use of the same UAV windows for both stages is ordinary multi-task supervised learning and does not force the result by construction. The large F1 gain is therefore an empirical claim open to external falsification, not a circular reduction.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 2 invented entities

The central F1 claim rests on standard ML practice plus domain rules for safety and uncertainty, a moderate measured correlation, and several free design choices (especially flip threshold τ and network sizes). No new physical entities; the invented pieces are algorithmic (GatedMLP uncertainty head, uLNR flip schedule). External validity hinges on the single Surrealist/PX4 simulation dataset and rule-based uncertainty labels from prior work.

free parameters (4)
  • flip threshold τ = 3.0
    Controls which safe windows are eligible for relabeling and the aggressiveness of flips; chosen as τ=3.0 after validation search over {0.5…3.5}; F1 is highly sensitive to this value (Table 4).
  • uncertainty predictor architecture dims (projection p, expansion e, dropout) = p=64, e=128, dropout=0.3
    Projection 64, expansion 128, dropout 0.3 selected by grid search; change the uncertainty scores that drive all flips.
  • safety predictor Bi-LSTM depth/hidden size and training hyperparameters = 3 layers, hidden 64, lr=1e-2
    3-layer Bi-LSTM hidden 64, AdamW lr=1e-2, etc., tuned on validation; affect absolute F1 of U-Balance and baselines.
  • window length N and channel set = N=25, C=4
    Fixed N=25 timesteps and channels (r,x,y,z) following prior UAV work; defines the sample unit for both predictors.
axioms (4)
  • domain assumption A flight window is unsafe iff UAV–obstacle distance falls below 1.5 m.
    Ground-truth safety labels (§2.1); all supervised metrics depend on this rule.
  • domain assumption Behavioral uncertainty labels can be generated automatically from kinematic inconsistencies (e.g., abrupt heading changes) as in Khatiri et al.
    Required to train the uncertainty predictor; procedure is outsourced to prior work (§3 intro).
  • ad hoc to paper High behavioral uncertainty among safe-labeled windows marks informative boundary samples worth flipping to the minority class.
    Core design premise of uLNR (§3.2); only partially supported by r_pb≈0.444 and ablations.
  • standard math Temporal order-preserving 8:1:1 split without future leakage is a valid evaluation protocol for this telemetry.
    Standard time-series ML practice (§4.2).
invented entities (2)
  • uLNR (uncertainty-guided label rebalancing) no independent evidence
    purpose: Map uncertainty z-scores of safe windows to Bernoulli flip probabilities via shifted tanh, enriching the unsafe class without synthesis.
    Algorithmic construct adapted from image LNR; no independent physical existence beyond this training procedure.
  • GatedMLP uncertainty predictor on distributional kinematic features no independent evidence
    purpose: Summarize each window into mean/std/min/max per channel and produce an uncertainty score used by uLNR.
    Architecture choice specific to this paper; ablations show it helps but it is not an external natural kind.

pith-pipeline@v1.1.0-grok45 · 28260 in / 3480 out tokens · 44181 ms · 2026-07-13T18:04:34.240620+00:00 · methodology

0 comments
read the original abstract

Safety monitoring is essential for Cyber-Physical Systems (CPSs). However, unsafe events are rare in real-world CPS operations, creating an extreme class imbalance that degrades safety predictors. Standard rebalancing techniques perform poorly on time-series CPS telemetry, either generating unrealistic synthetic samples or overfitting on the minority class. Meanwhile, behavioral uncertainty in CPS operations, defined as the degree of doubt or uncertainty in CPS decisions , is often correlated with safety outcomes but unexplored in safety monitoring. To that end, we propose U-Balance, a supervised approach that leverages behavioral uncertainty to rebalance imbalanced datasets prior to training a safety predictor. U-Balance first trains a GatedMLP-based uncertainty predictor that summarizes each telemetry window into distributional kinematic features and outputs an uncertainty score. It then applies an uncertainty-guided label rebalancing (uLNR) mechanism that probabilistically relabels $\textit{safe}$-labeled windows with unusually high uncertainty as $\textit{unsafe}$, thereby enriching the minority class with informative boundary samples without synthesizing new data. Finally, a safety predictor is trained on the rebalanced dataset for safety monitoring. We evaluate U-Balance on a large-scale UAV benchmark with a 46:1 safe-to-unsafe ratio. Results confirm a moderate but significant correlation between behavioral uncertainty and safety. We then identify uLNR as the most effective strategy to exploit uncertainty information, compared to direct early and late fusion. U-Balance achieves a 0.806 F1 score, outperforming the strongest baseline by 14.3 percentage points, while maintaining competitive inference efficiency. Ablation studies confirm that both the GatedMLP-based uncertainty predictor and the uLNR mechanism contribute significantly to U-Balance's effectiveness.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

94 extracted references · 5 linked inside Pith

  1. [1]

    Moloud Abdar, Farhad Pourpanah, Sadiq Hussain, Dana Rezazadegan, Li Liu, Mo- hammad Ghavamzadeh, Paul Fieguth, Xiaochun Cao, Abbas Khosravi, U Rajendra Acharya, et al. 2021. A review of uncertainty quantification in deep learning: Techniques, applications and challenges.Information fusion76 (2021), 243–297

    2021

  2. [2]

    Marian Sorin Adam, Morten Larsen, Kjeld Jensen, and Ulrik Pagh Schultz. 2016. Rule-based dynamic safety monitoring for mobile robots.Journal of Software Engineering for Robotics7, 1 (2016), 120–141

    2016

  3. [3]

    Sorin Adam, Morten Larsen, Kjeld Jensen, and Ulrik Pagh Schultz. 2014. Towards rule-based dynamic safety monitoring for mobile robots. InInternational Confer- ence on Simulation, Modeling, and Programming for Autonomous Robots. Springer, 207–218

    2014

  4. [4]

    Andrea Arcuri and Lionel Briand. 2011. A practical guide for using statistical tests to assess randomized algorithms in software engineering. InProceedings of the 33rd International Conference on Software Engineering(Waikiki, Honolulu, HI, USA)(ICSE ’11). Association for Computing Machinery, New York, NY, USA, 1–10. doi:10.1145/1985793.1985795

    doi:10.1145/1985793.1985795 2011

  5. [5]

    Batuhan Bakirarar and Atilla Halil Elhan. 2023. Class weighting technique to deal with imbalanced class problem in machine learning: Methodological research. Türkiye Klinikleri Biyoistatistik15, 1 (2023), 19–29

    2023

  6. [6]

    Gustavo EAPA Batista, Andre CPLF Carvalho, and Maria Carolina Monard. 2000. Applying one-sided selection to unbalanced datasets. InMexican International Conference on Artificial Intelligence. Springer, 315–325

    2000

  7. [7]

    Kirstie Bellman, Jean Botev, Ada Diaconescu, Lukas Esterle, Christian Gruhl, Christopher Landauer, Peter R Lewis, Phyllis R Nelson, Evangelos Pournaras, An- thony Stein, et al. 2021. Self-improving system integration: Mastering continuous change.Future Generation Computer Systems117 (2021), 29–46

    2021

  8. [8]

    2006.Pattern recognition and machine learning

    Christopher M Bishop and Nasser M Nasrabadi. 2006.Pattern recognition and machine learning. Vol. 4. Springer

    2006

  9. [9]

    Said Yacine Boulahia, Abdenour Amamra, Mohamed Ridha Madi, and Said Daikh

  10. [10]

    Early, intermediate and late fusion strategies for robust deep learning-based multimodal action recognition.Machine Vision and Applications32, 6 (2021), 121

    2021

  11. [11]

    Leo Breiman. 2001. Random forests.Machine learning45, 1 (2001), 5–32

    2001

  12. [12]

    Kaidi Cao, Colin Wei, Adrien Gaidon, Nikos Arechiga, and Tengyu Ma. 2019. Learning imbalanced datasets with label-distribution-aware margin loss.Ad- vances in neural information processing systems32 (2019)

    2019

  13. [13]

    Ferhat Ozgur Catak, Tao Yue, and Shaukat Ali. 2022. Uncertainty-aware pre- diction validator in deep learning models for cyber-physical system data.ACM Transactions on Software Engineering and Methodology (TOSEM)31, 4 (2022), 1–31

    2022

  14. [14]

    Varun Chandola, Arindam Banerjee, and Vipin Kumar. 2009. Anomaly detection: A survey.ACM computing surveys (CSUR)41, 3 (2009), 1–58

    2009

  15. [15]

    Nitesh V Chawla, Kevin W Bowyer, Lawrence O Hall, and W Philip Kegelmeyer

  16. [16]

    SMOTE: synthetic minority over-sampling technique.Journal of artificial intelligence research16 (2002), 321–357

    2002

  17. [17]

    Wuxing Chen, Kaixiang Yang, Zhiwen Yu, Yifan Shi, and CL Philip Chen. 2024. A survey on imbalanced learning: latest research, applications and future directions. Artificial Intelligence Review57, 6 (2024), 137

    2024

  18. [18]

    Kyunghyun Cho, Bart Van Merriënboer, Çağlar Gulçehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. 2014. Learning phrase representations using RNN encoder–decoder for statistical machine translation. InProceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). 1724–1734

    2014

  19. [19]

    Hsin-Ping Chou, Shih-Chieh Chang, Jia-Yu Pan, Wei Wei, and Da-Cheng Juan

  20. [20]

    InEuropean conference on computer vision

    Remix: rebalanced mixup. InEuropean conference on computer vision. Springer, 95–110

  21. [21]

    Thomas Cover and Peter Hart. 1967. Nearest neighbor pattern classification. IEEE transactions on information theory13, 1 (1967), 21–27

    1967

  22. [22]

    David R Cox. 1958. The regression analysis of binary sequences.Journal of the Royal Statistical Society Series B: Statistical Methodology20, 2 (1958), 215–232

    1958

  23. [23]

    1986.Introduction to classical and modern test theory.ERIC

    Linda Crocker and James Algina. 1986.Introduction to classical and modern test theory.ERIC

    1986

  24. [24]

    Raffaello D’Andrea. 2014. Guest editorial can drones deliver?IEEE Transactions on Automation Science and Engineering11, 3 (2014), 647–648

    2014

  25. [25]

    Rahul Dey and Fathi M Salem. 2017. Gate-variants of gated recurrent unit (GRU) neural networks. In2017 IEEE 60th international midwest symposium on circuits and systems (MWSCAS). IEEE, 1597–1600

    2017

  26. [26]

    Andrea Di Sorbo, Fiorella Zampetti, Aaron Visaggio, Massimiliano Di Penta, and Sebastiano Panichella. 2023. Automated identification and qualitative characteri- zation of safety concerns reported in uav software platforms.ACM Transactions on Software Engineering and Methodology32, 3 (2023), 1–37

    2023

  27. [27]

    Alexandre Donzé and Oded Maler. 2010. Robust satisfaction of temporal logic over real-valued signals. InInternational conference on formal modeling and analysis of timed systems. Springer, 92–106

    2010

  28. [28]

    Wenli Duo, MengChu Zhou, and Abdullah Abusorrah. 2022. A survey of cyber attacks on cyber physical systems: Recent advances and challenges.IEEE/CAA Journal of Automatica Sinica9, 5 (2022), 784–800

    2022

  29. [29]

    Charles Elkan. 2001. The foundations of cost-sensitive learning. InInternational joint conference on artificial intelligence, Vol. 17. Lawrence Erlbaum Associates Ltd, 973–978

    2001

  30. [30]

    Manuel Fernández-Delgado, Eva Cernadas, Senén Barro, and Dinani Amorim

  31. [31]

    Do we need hundreds of classifiers to solve real world classification problems?The journal of machine learning research15, 1 (2014), 3133–3181

    2014

  32. [32]

    Jerome H Friedman. 2001. Greedy function approximation: a gradient boosting machine.Annals of statistics(2001), 1189–1232

    2001

  33. [33]

    Konrad Gadzicki, Razieh Khamsehashari, and Christoph Zetzsche. 2020. Early vs late fusion in multimodal convolutional neural networks. In2020 IEEE 23rd international conference on information fusion (FUSION). IEEE, 1–6

    2020

  34. [34]

    Ben Grindley, Katie Phillips, Katie J Parnell, Tom Cherrett, James Scanlan, and Katherine L Plant. 2024. Over a decade of UAV incidents: A human factors analysis of causal factors.Applied Ergonomics121 (2024), 104355

    2024

  35. [35]

    Hui Han, Wen-Yuan Wang, and Bing-Huan Mao. 2005. Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning. InInternational conference on intelligent computing. Springer, 878–887

    2005

  36. [36]

    Liping Han, Shaukat Ali, Tao Yue, Aitor Arrieta, and Maite Arratibel. 2023. Uncertainty-aware robustness assessment of industrial elevator systems.ACM Transactions on Software Engineering and Methodology32, 4 (2023), 1–51

    2023

  37. [37]

    2022.Investigation of UA V related incidents and accidents

    Sigurd Haugse. 2022.Investigation of UA V related incidents and accidents. Master’s thesis. UiT The Arctic University of Norway

    2022

  38. [38]

    Haibo He, Yang Bai, Edwardo A Garcia, and Shutao Li. 2008. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelli- gence). Ieee, 1322–1328

    2008

  39. [39]

    Mohammad Hossin and Md Nasir Sulaiman. 2015. A review on evaluation metrics for data classification evaluations.International journal of data mining & knowledge management process5, 2 (2015), 1

    2015

  40. [40]

    Guangzheng Hu, Feng Liu, Mingming Gong, Guanghui Wang, and Liuhua Peng

  41. [41]

    InProceedings of the 42nd International Conference on Machine Learning (Proceedings of Machine Conference acronym ’XX, June 03–05, 2018, Woodstock, NY Ayotunde et al

    Learning Imbalanced Data with Beneficial Label Noise. InProceedings of the 42nd International Conference on Machine Learning (Proceedings of Machine Conference acronym ’XX, June 03–05, 2018, Woodstock, NY Ayotunde et al. Learning Research, Vol. 267), Aarti Singh, Maryam Fazel, Daniel Hsu, Simon Lacoste-Julien, Felix Berkenkamp, Tegan Maharaj, Kiri Wagstaf...

    2018

  42. [42]

    Zhiheng Huang, Wei Xu, and Kai Yu. 2015. Bidirectional LSTM-CRF models for sequence tagging.arXiv preprint arXiv:1508.01991(2015)

    Pith/arXiv arXiv 2015

  43. [43]

    Nicholas Jeffrey, Qing Tan, and José R Villar. 2024. Using ensemble learning for anomaly detection in cyber–physical systems.Electronics13, 7 (2024), 1391

    2024

  44. [44]

    Yuchen Jiang, Shen Yin, and Okyay Kaynak. 2018. Data-driven monitoring and safety control of industrial cyber-physical systems: Basics and beyond.IEEe Access6 (2018), 47374–47384

    2018

  45. [45]

    Justin M Johnson and Taghi M Khoshgoftaar. 2019. Survey on deep learning with class imbalance.Journal of big data6, 1 (2019), 27

    2019

  46. [46]

    Piotr Jan Kasprzyk and Anna Konert. 2021. Reporting and investigation of Unmanned Aircraft Systems (UAS) accidents and serious incidents. Regulatory perspective.Journal of Intelligent & Robotic Systems103, 1 (2021), 3

    2021

  47. [47]

    Salman H Khan, Munawar Hayat, Mohammed Bennamoun, Ferdous A Sohel, and Roberto Togneri. 2017. Cost-sensitive learning of deep feature representations from imbalanced data.IEEE transactions on neural networks and learning systems 29, 8 (2017), 3573–3587

    2017

  48. [48]

    Sajad Khatiri, Fatemeh Mohammadi Amin, Sebastiano Panichella, and Paolo Tonella. 2025. When uncertainty leads to unsafety: Empirical insights into the role of uncertainty in unmanned aerial vehicle safety.Empirical Software Engineering30, 6 (2025), 166

    2025

  49. [49]

    Sajad Khatiri, Sebastiano Panichella, and Paolo Tonella. 2023. Simulation-based Test Case Generation for Unmanned Aerial Vehicles in the Neighborhood of Real Flights. In2023 IEEE Conference on Software Testing, Verification and Validation (ICST). 281–292. doi:10.1109/ICST57152.2023.00034

    doi:10.1109/icst57152.2023.00034 2023

  50. [50]

    Sajad Khatiri, Sebastiano Panichella, and Paolo Tonella. 2023. Simulation-based test case generation for unmanned aerial vehicles in the neighborhood of real flights. In2023 IEEE Conference on Software Testing, Verification and Validation (ICST). IEEE, 281–292

    2023

  51. [51]

    Diana Kornbrot. 2014. Point biserial correlation.Wiley StatsRef: Statistics Reference Online(2014)

    2014

  52. [52]

    Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. 2002. Gradient- based learning applied to document recognition.Proc. IEEE86, 11 (2002), 2278– 2324

    2002

  53. [53]

    Nancy Leveson, Chris Wilkinson, Cody Fleming, John Thomas, and Ian Tracy

  54. [54]

    Massachusetts Institute of Technology, Cambridge, MA(2014)

    A comparison of STPA and the ARP 4761 safety assessment process. Massachusetts Institute of Technology, Cambridge, MA(2014)

    2014

  55. [55]

    Mengke Li, Yiu-ming Cheung, and Yang Lu. 2022. Long-tailed visual recognition via gaussian clouded logit adjustment. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition. 6929–6938

    2022

  56. [56]

    Bryan Lim, Sercan Ö Arık, Nicolas Loeff, and Tomas Pfister. 2021. Temporal fusion transformers for interpretable multi-horizon time series forecasting.International journal of forecasting37, 4 (2021), 1748–1764

    2021

  57. [57]

    Vivian Lin, Ramneet Kaur, Yahan Yang, Souradeep Dutta, Yiannis Kantaros, Anirban Roy, Susmit Jha, Oleg Sokolsky, and Insup Lee. 2025. Safety monitoring for learning-enabled cyber-physical systems in out-of-distribution scenarios. In Proceedings of the ACM/IEEE 16th International Conference on Cyber-Physical Systems (with CPS-IoT Week 2025). 1–11

    2025

  58. [58]

    J MacQueen. 1967. Multivariate observations. InProceedings ofthe 5th Berke- ley symposium on mathematical statisticsand probability, Vol. 1. University of California press Oakland, CA, USA, 281–297

    1967

  59. [59]

    Oded Maler and Dejan Nickovic. 2004. Monitoring temporal properties of con- tinuous signals. InInternational symposium on formal techniques in real-time and fault-tolerant systems. Springer, 152–166

    2004

  60. [60]

    Rhiannon Michelmore, Matthew Wicker, Luca Laurenti, Luca Cardelli, Yarin Gal, and Marta Kwiatkowska. 2020. Uncertainty quantification with statistical guarantees in end-to-end autonomous driving control. In2020 IEEE international conference on robotics and automation (ICRA). IEEE, 7344–7350

    2020

  61. [61]

    Ojas Ankush Naik, Pooja Shyamsundar, and Navrati Saxena. 2025. AI-Enabled Anticipatory Handover Predictions in 5G Networks. In2025 International Confer- ence on Emerging Trends in Networks and Computer Communications (ETNCC). IEEE, 61–68

    2025

  62. [62]

    Vinod Nair and Geoffrey E. Hinton. 2010. Rectified linear units improve restricted boltzmann machines. InProceedings of the 27th International Conference on Inter- national Conference on Machine Learning(Haifa, Israel)(ICML’10). Omnipress, Madison, WI, USA, 807–814

    2010

  63. [63]

    Bhushankumar Nemade, Kiran Kishor Maharana, Vikram Kulkarni, Surajit Mon- dal, GS Pradeep Ghantasala, Amal Al-Rasheed, Masresha Getahun, and Ben Oth- man Soufiene. 2024. IoT-based automated system for water-related disease prediction.Scientific Reports14, 1 (2024), 29483

    2024

  64. [64]

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. 2019. Pytorch: An imperative style, high-performance deep learning library.Advances in neural information processing systems32 (2019)

    2019

  65. [65]

    Ross Quinlan

    J. Ross Quinlan. 1986. Induction of decision trees.Machine learning1, 1 (1986), 81–106

    1986

  66. [66]

    Shrinivas Ramasubramanian, Harsh Rangwani, Sho Takemori, Kunal Samanta, Yuhei Umeda, and Venkatesh Babu Radhakrishnan. 2024. Selective Mixup Fine-Tuning for Optimizing Non-Decomposable Objectives.arXiv preprint arXiv:2403.18301(2024)

    Pith/arXiv arXiv 2024

  67. [67]

    David E Rumelhart, Geoffrey E Hinton, and Ronald J Williams. 1986. Learning representations by back-propagating errors.nature323, 6088 (1986), 533–536

    1986

  68. [68]

    Bhargava Satya, Rama Krishna Gorthi, et al. 2024. Effective-LDAM: An Effective Loss Function To Mitigate Data Imbalance for Robust Chest X-Ray Disease Classification.arXiv preprint arXiv:2407.04953(2024)

    Pith/arXiv arXiv 2024

  69. [70]

    Sepehr Sharifi, Andrea Stocco, and Lionel C Briand. 2025. System safety monitor- ing of learned components using temporal metric forecasting.ACM Transactions on Software Engineering and Methodology34, 6 (2025), 1–43

    2025

  70. [71]

    Sanjiv Sharma and D Chakravarti. 2005. UAV operations: An analysis of incidents and accidents with human factors and crew resource management perspective. Indian Journal of Aerospace Medicine49, 1 (2005), 29–36

    2005

  71. [72]

    Xiaoming Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, and Ming Jin. 2024. Time-moe: Billion-scale time series foundation models with mixture of experts.arXiv preprint arXiv:2409.16040(2024)

    Pith/arXiv arXiv 2024

  72. [73]

    Sima Siami-Namini, Neda Tavakoli, and Akbar Siami Namin. 2019. The perfor- mance of LSTM and BiLSTM in forecasting time series. In2019 IEEE International conference on big data (Big Data). IEEE, 3285–3292

    2019

  73. [74]

    Rohan Sinha, Edward Schmerling, and Marco Pavone. 2023. Closing the loop on runtime monitors with fallback-safe mpc. In2023 62nd IEEE Conference on Decision and Control (CDC). IEEE, 6533–6540

    2023

  74. [75]

    Cees G. M. Snoek, Marcel Worring, and Arnold W. M. Smeulders. 2005. Early versus late fusion in semantic video analysis. InProceedings of the 13th Annual ACM International Conference on Multimedia(Hilton, Singapore)(MULTIMEDIA ’05). Association for Computing Machinery, New York, NY, USA, 399–402. doi:10. 1145/1101149.1101236

    arXiv 2005

  75. [76]

    2008.Support vector machines

    Ingo Steinwart and Andreas Christmann. 2008.Support vector machines. Springer Science & Business Media

    2008

  76. [77]

    Andrea Stocco, Michael Weiss, Marco Calzana, and Paolo Tonella. 2020. Mis- behaviour prediction for autonomous driving systems. InProceedings of the ACM/IEEE 42nd international conference on software engineering. 359–371

    2020

  77. [78]

    Shuai-Wen Tang, Zhi-Jie Zhou, Chang-Hua Hu, Fu-Jun Zhao, and You Cao. 2020. A new evidential reasoning rule-based safety assessment method with sensor reliability for complex systems.IEEE Transactions on Cybernetics52, 5 (2020), 4027–4038

    2020

  78. [79]

    Seema Varma. 2006. Preliminary item statistics using point-biserial correlation and p-values.Educational Data Systems Inc.: Morgan Hill CA. Retrieved16, 07 (2006), 1–7

    2006

  79. [80]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need.Advances in neural information processing systems30 (2017)

    2017

  80. [81]

    Jianfeng Wang, Thomas Lukasiewicz, Xiaolin Hu, Jianfei Cai, and Zhenghua Xu

Showing first 80 references.