mlsanityData to 14 Oct 2026

Learning Perturbations to Explain Time Series Predictions

arXiv 2305.18840 ↗First posted 30 May 2023cs.LG cs.AI stat.MLcited by 23

Not on the map. Why?

Abstract (arXiv)

Explaining predictions based on multivariate time series data carries the additional difficulty of handling not only multiple features, but also time dependencies. It matters not only what happened, but also when, and the same feature could have a very different impact on a prediction depending on this time information. Previous work has used perturbation-based saliency methods to tackle this issue, perturbing an input using a trainable mask to discover which features at which times are driving the predictions. However these methods introduce fixed perturbations, inspired from similar methods on static data, while there seems to be little motivation to do so on temporal data. In this work, we aim to explain predictions by learning not only masks, but also associated perturbations. We empirically show that learning these perturbations significantly improves the quality of these explanations on time series data.

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). Builds on 13 of its 23 references that are in the map. First posted on arXiv 30 May 2023.

Builds on

13 of 23 references.

References · 13

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  3. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  5. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  6. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  7. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,304
  8. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 612
  9. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 530
  10. 2019A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAIcited by 404
  11. 2016RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanismcited by 349
  12. 2019Understanding Deep Networks via Extremal Perturbations and Smooth Maskscited by 259
1 more
  1. 2021Explaining Time Series Predictions with Dynamic Maskscited by 73

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 5; 2025: 6; 2026: 11.

23 papers.

Sort:

By area:

  1. 2024TimeX++: Learning Time-Series Explanations with Information Bottleneckcited by 33
  2. 2024Explaining Time Series via Contrastive and Locally Sparse Perturbationscited by 22
  3. 2025Deep Learning for Time Series Forecasting: A Surveycited by 10
  4. 2025TIMING: Temporality-Aware Integrated Gradients for Time Series Explanationcited by 10
  5. 2024FLEXtime: Filterbank learning to explain time seriescited by 6
  6. 2023Time Interpret: a Unified Model Interpretability Library for Time Seriescited by 4
  7. 2024FreqRISE: Explaining time series using frequency maskingcited by 4
  8. 2025DeltaSHAP: Explaining Prediction Evolutions in Online Patient Monitoring with Shapley Valuescited by 2
  9. 2026Dissecting Chronos: Sparse Autoencoders Reveal Causal Feature Hierarchies in Time Series Foundation Modelscited by 2
  10. 2025Failure Modes of Time Series Interpretability Algorithms for Critical Care Applications and Potential Solutionscited by 1
  11. 2026xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truthcited by 1
  12. 2026Information Bottleneck Learning for Faithful Time Series Forecasting Explanationscited by 1
11 more
  1. 2024WinTSR: A Windowed Temporal Saliency Rescaling Method for Interpreting Time Series Deep Learning Modelscited by 0
  2. 2025Delta-XAI: A Unified Framework for Explaining Prediction Changes in Online Time Series Monitoringcited by 0
  3. 2025When, How Long and How Much? Interpretable Neural Networks for Time Series Regression by Learning to Mask and Aggregatecited by 0
  4. 2026TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Modelscited by 0
  5. 2026Turning Time Series into Algebraic Equations: Symbolic Machine Learning for Interpretable Modeling of Chaotic Time Seriescited by 0
  6. 2026Explainable AI Using Inherently Interpretable Components for Wearable-based Health Monitoringcited by 0
  7. 2026Global Explanations for Multivariate Time Series Forecasting Models via K-Order Markov Approximationscited by 0
  8. 2026Beyond Sufficiency: Time Series Explanation with Counterfactual Necessitycited by 0
  9. 2026Towards A Unified Information Bottleneck Framework for Time Series Explanationscited by 0
  10. 2026Explaining Time Series Forecasting with Horizon-Resolved Attributioncited by 0
  11. 2026Learnable Time-Frequency Masks for Explaining Time-Series Classifierscited by 0