mlsanityData to 7 Oct 2026

Explaining Time Series via Contrastive and Locally Sparse Perturbations

arXiv 2401.08552 ↗First posted 16 Jan 2024cs.LG cs.AIcited by 22

Not on the map. Why?

Abstract (arXiv)

Explaining multivariate time series is a compound challenge, as it requires identifying important locations in the time series and matching complex temporal patterns. Although previous saliency-based methods addressed the challenges, their perturbation may not alleviate the distribution shift issue, which is inevitable especially in heterogeneous samples. We present ContraLSP, a locally sparse model that introduces counterfactual samples to build uninformative perturbations but keeps distribution using contrastive learning. Furthermore, we incorporate sample-specific sparse gates to generate more binary-skewed and smooth masks, which easily integrate temporal trends and select the salient features parsimoniously. Empirical studies on both synthetic and real-world datasets show that ContraLSP outperforms state-of-the-art models, demonstrating a substantial improvement in explanation quality for time series data. The source code is available at urlhttps://github.com/zichuan-liu/ContraLSP.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Builds on 36 of its 51 references that are in the map. First posted on arXiv 16 Jan 2024.

Builds on

36 of 51 references.

References · 36

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  2. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  3. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  4. 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,833
  5. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  6. 2017Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learningcited by 935
  7. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 611
  8. 2017Learning Sparse Neural Networks through L0 Regularizationcited by 571
  9. 2019A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAIcited by 403
  10. 2016RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanismcited by 344
  11. 2019Counterfactual Visual Explanationscited by 312
  12. 2019Understanding Deep Networks via Extremal Perturbations and Smooth Maskscited by 257
24 more
  1. 2020Benchmarking Deep Learning Interpretability in Time Series Predictionscited by 102
  2. 2016Group Sparse Regularization for Deep Neural Networkscited by 89
  3. 2020Evaluating Saliency Map Explanations for Convolutional Neural Networks: A User Studycited by 86
  4. 2021The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanationscited by 86
  5. 2020TimeSHAP: Explaining Recurrent Models through Sequence Perturbationscited by 73
  6. 2021Explaining Time Series Predictions with Dynamic Maskscited by 72
  7. 2019Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attentioncited by 71
  8. 2021Explainable Artificial Intelligence (XAI) on TimeSeries Data: A Surveycited by 71
  9. 2020Learning What Makes a Difference from Counterfactual Examples and Gradient Supervisioncited by 66
  10. 2020Instance-based Counterfactual Explanations for Time Series Classificationcited by 64
  11. 2020Evaluations and Methods for Explanation through Robustness Analysiscited by 51
  12. 2021OOD-CV: A Benchmark for Robustness to Out-of-Distribution Shifts of Individual Nuisances in Natural Imagescited by 46
  13. 2023Encoding Time-Series Explanations through Self-Supervised Model Behavior Consistencycited by 42
  14. 2017Sparse-Input Neural Networks for High-dimensional Nonparametric Regression and Classificationcited by 39
  15. 2021Locally Sparse Neural Networks for Tabular Biomedical Datacited by 34
  16. 2022ProtoVAE: A Trustworthy Self-Explainable Prototypical Variational Modelcited by 34
  17. 2020Preserving Dynamic Attention for Long-Term Spatial-Temporal Predictioncited by 30
  18. 2021Temporal Dependencies in Feature Importance for Time Series Predictionscited by 27
  19. 2023Learning Perturbations to Explain Time Series Predictionscited by 23
  20. 2019Input-Cell Attention Reduces Vanishing Saliency of Recurrent Neural Networkscited by 20
  21. 2023CoRTX: Contrastive Framework for Real-time Explanationcited by 20
  22. 2023NA2Q: Neural Attention Additive Model for Interpretable Multi-Agent Q-Learningcited by 14
  23. 2016Localized Lasso for High-Dimensional Regressioncited by 11
  24. 2023Don't be fooled: label leakage in explanation methods and the importance of their quantitative evaluationcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 7; 2026: 10.

22 papers.

Sort:

By area:

  1. 2024TimeX++: Learning Time-Series Explanations with Information Bottleneckcited by 32
  2. 2024Protecting Your LLMs with Information Bottleneckcited by 26
  3. 2025TIMING: Temporality-Aware Integrated Gradients for Time Series Explanationcited by 9
  4. 2025Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedbackcited by 7
  5. 2024FLEXtime: Filterbank learning to explain time seriescited by 6
  6. 2024FreqRISE: Explaining time series using frequency maskingcited by 4
  7. 2025ShapeX: Shapelet-Driven Post Hoc Explanations for Time Series Classification Modelscited by 3
  8. 2025Time-series attribution maps with regularized contrastive learningcited by 2
  9. 2025DeltaSHAP: Explaining Prediction Evolutions in Online Patient Monitoring with Shapley Valuescited by 2
  10. 2026Diversified Scaling Inference in Time Series Foundation Modelscited by 2
  11. 2025TriShGAN: Enhancing Sparsity and Robustness in Multivariate Time Series Counterfactuals Explanationcited by 1
  12. 2026Dissecting Chronos: Sparse Autoencoders Reveal Causal Feature Hierarchies in Time Series Foundation Modelscited by 1
10 more
  1. 2026Information Bottleneck Learning for Faithful Time Series Forecasting Explanationscited by 1
  2. 2024WinTSR: A Windowed Temporal Saliency Rescaling Method for Interpreting Time Series Deep Learning Modelscited by 0
  3. 2025Delta-XAI: A Unified Framework for Explaining Prediction Changes in Online Time Series Monitoringcited by 0
  4. 2026TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Modelscited by 0
  5. 2026Explainable AI Using Inherently Interpretable Components for Wearable-based Health Monitoringcited by 0
  6. 2026Global Explanations for Multivariate Time Series Forecasting Models via K-Order Markov Approximationscited by 0
  7. 2026Feature Attribution-Based Explainability Analysis of Deep Learning Models in Predictive Process Monitoringcited by 0
  8. 2026Towards A Unified Information Bottleneck Framework for Time Series Explanationscited by 0
  9. 2026Explaining Time Series Forecasting with Horizon-Resolved Attributioncited by 0
  10. 2026Learnable Time-Frequency Masks for Explaining Time-Series Classifierscited by 0