mlsanityData to 7 Oct 2026

Evaluation of post-hoc interpretability methods in time-series classification

arXiv 2202.05656 ↗First posted 11 Feb 2022cs.LG cs.AI stat.MLcited by 53

Not on the map. Why?

Abstract (arXiv)

Post-hoc interpretability methods are critical tools to explain neural-network results. Several post-hoc methods have emerged in recent years, but when applied to a given task, they produce different results, raising the question of which method is the most suitable to provide correct post-hoc interpretability. To understand the performance of each method, quantitative evaluation of interpretability methods is essential. However, currently available frameworks have several drawbacks which hinders the adoption of post-hoc interpretability methods, especially in high-risk sectors. In this work, we propose a framework with quantitative metrics to assess the performance of existing post-hoc interpretability methods in particular in time series classification. We show that several drawbacks identified in the literature are addressed, namely dependence on human judgement, retraining, and shift in the data distribution when occluding samples. We additionally design a synthetic dataset with known discriminative features and tunable complexity. The proposed methodology and quantitative metrics can be used to understand the reliability of interpretability methods results obtained in practical applications. In turn, they can be embedded within operational workflows in critical fields that require accurate interpretability results for e.g., regulatory policies.

Key facts. Cited by 53 papers in the map (data through 7 Oct 2026). Builds on 28 of its 57 references that are in the map. First posted on arXiv 11 Feb 2022.

Builds on

28 of 57 references.

References · 28

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  5. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  6. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  7. 2018Sanity Checks for Saliency Mapscited by 1,149
  8. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 748
  9. 2016The Mythos of Model Interpretabilitycited by 716
  10. 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 590
  11. 2018Deep learning for time series classification: a reviewcited by 575
  12. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 542
16 more
  1. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 467
  2. 2018The UCR Time Series Archivecited by 421
  3. 2015Evaluating the visualization of what a Deep Neural Network has learnedcited by 416
  4. 2018Scalable and accurate deep learning for electronic health recordscited by 275
  5. 2018DropBlock: A regularization method for convolutional networkscited by 261
  6. 2015Explaining NonLinear Classification Decisions with Deep Taylor Decompositioncited by 248
  7. 2019Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directionscited by 172
  8. 2020Benchmarking Deep Learning Interpretability in Time Series Predictionscited by 102
  9. 2020Improving data-driven global weather prediction using deep convolutional neural networks on a cubed spherecited by 95
  10. 2021Voice2Series: Reprogramming Acoustic Models for Time Series Classificationcited by 91
  11. 2019Towards a Rigorous Evaluation of XAI Methods on Time Seriescited by 67
  12. 2019Let's Agree to Agree: Neural Networks Share Classification Order on Real Datasetscited by 55
  13. 2021Do Input Gradients Highlight Discriminative Features?cited by 52
  14. 2018Feature Space Transfer for Data Augmentationcited by 21
  15. 2020Deep Learning for Road Traffic Forecasting: Does it Make a Difference?cited by 21
  16. 2021Medical Imaging and Machine Learningcited by 12

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 8; 2024: 10; 2025: 20; 2026: 14.

53 papers.

Sort:

By area:

  1. 2023A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethicscited by 162
  2. 2023A Comprehensive Review on Financial Explainable AIcited by 10
  3. 2024Explainable Natural Language Processing for Corporate Sustainability Analysiscited by 9
  4. 2024Large Language Models in Plant Biologycited by 8
  5. 2024ProtoS-ViT: Visual foundation models for sparse self-explainable classificationscited by 8
  6. 2023Interpretable machine learning for time-to-event prediction in medicine and healthcarecited by 7
  7. 2023Sparse learned kernels for interpretable and efficient medical time series processingcited by 6
  8. 2024Explain the Black Box for the Sake of Science: the Scientific Method in the Era of Generative Artificial Intelligencecited by 5
  9. 2024Improving the Evaluation and Actionability of Explanation Methods for Multivariate Time Series Classificationcited by 5
  10. 2024Revisiting the robustness of post-hoc interpretability methodscited by 5
  11. 2025Class-Dependent Perturbation Effects in Evaluating Time Series Attributionscited by 5
  12. 2022Don't Get Me Wrong: How to Apply Deep Visual Interpretations to Time Seriescited by 4
38 more
  1. 2023Enhancing Interpretability and Generalizability in Extended Isolation Forestscited by 4
  2. 2024What Sketch Explainability Really Means for Downstream Taskscited by 4
  3. 2024CAFO: Feature-Centric Explanation on Time Series Classificationcited by 4
  4. 2025ProtoECGNet: Case-Based Interpretable Deep Learning for Multi-Label ECG Classification with Contrastive Learningcited by 4
  5. 2026Explainable AI: Learning from the Learnerscited by 4
  6. 2025Transparent AI: The Case for Interpretability and Explainabilitycited by 3
  7. 2023Explainable AI for clinical risk prediction: a survey of concepts, methods, and modalitiescited by 2
  8. 2023Evaluating Explanation Methods for Multivariate Time Series Classificationcited by 2
  9. 2024Quantum Adjoint Convolutional Layers for Effective Data Representationcited by 2
  10. 2024MetaWearS: A Shortcut in Wearable Systems Lifecycle with Only a Few Shotscited by 2
  11. 2025Tell me why: Visual foundation models as self-explainable classifierscited by 2
  12. 2025XAI4Extremes: An interpretable machine learning framework for understanding extreme-weather precursors under climate changecited by 2
  13. 2025Why Do Class-Dependent Evaluation Effects Occur with Time Series Feature Attributions? A Synthetic Data Investigationcited by 2
  14. 2026AnchorMoE: Interpretable Time Series Classification via Anchor-Routed MoEcited by 2
  15. 2023FinXABSA: Explainable Finance through Aspect-Based Sentiment Analysiscited by 1
  16. 2025Exploring Neural Granger Causality with xLSTMs: Unveiling Temporal Dependencies in Complex Datacited by 1
  17. 2025Beyond Explainability: The Case for AI Validationcited by 1
  18. 2025An Empirical Evaluation of Factors Affecting SHAP Explanation of Time Series Classificationcited by 1
  19. 2025Explainability matters: The effect of liability rules on the healthcare sectorcited by 1
  20. 2026xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truthcited by 1
  21. 2026One Step Closer to Ground Truth: A Multi-Scale Residual-Aware Representation Learning Pipeline for Predicting Time Series Datacited by 1
  22. 2025Adaptive Law-Based Transformation (ALT): A Lightweight Feature Representation for Time Series Classificationcited by 0
  23. 2025Feature Importance Depends on Properties of the Data: Towards Choosing the Correct Explanations for Your Data and Decision Trees based Modelscited by 0
  24. 2025Are We Merely Justifying Results ex Post Facto? Quantifying Explanatory Inversion in Post-Hoc Model Explanationscited by 0
  25. 2025ClimaEmpact: Domain-Aligned Small Language Models and Datasets for Extreme Weather Analyticscited by 0
  26. 2025CanadaFireSat: Toward high-resolution wildfire forecasting with multiple modalitiescited by 0
  27. 2025Towards Interpretable Concept Learning over Time Series via Temporal Logic Semanticscited by 0
  28. 2025DeepACTIF: Efficient Feature Attribution via Activation Traces in Neural Sequence Modelscited by 0
  29. 2025Learning Temporal Saliency for Time Series Forecasting with Cross-Scale Attentioncited by 0
  30. 2025Explainable Heterogeneous Anomaly Detection in Financial Networks via Adaptive Expert Routingcited by 0
  31. 2025Function Based Isolation Forest (FuBIF): A Unifying Framework for Interpretable Isolation-Based Anomaly Detectioncited by 0
  32. 2026TimeSliver : Symbolic-Linear Decomposition for Explainable Time Series Classificationcited by 0
  33. 2026Interpretability in Deep Time Series Models Demands Semantic Alignmentcited by 0
  34. 2026OmniSapiens: A Foundation Model for Social Behavior Processing via Heterogeneity-Aware Relative Policy Optimizationcited by 0
  35. 2026Hierarchical Industrial Demand Forecasting with Temporal and Uncertainty Explanationscited by 0
  36. 2026Does Explanation Correctness Matter? Linking Computational XAI Evaluation to Human Understandingcited by 0
  37. 2026AIM: Adversarial Information Masking for Faithfulness Evaluation of Saliency Mapscited by 0
  38. 2026Scaling Time Series Classification via XAI-Driven Data Reductioncited by 0

50 most cited of 53 All 53 (JSON)