# Learning Perturbations to Explain Time Series Predictions

arXiv: https://arxiv.org/abs/2305.18840 · HTML: https://mlsanity.com/paper/2305.18840/

## 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 (most cited first, 13 of 13)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Learning Important Features Through Propagating Activation Differences](https://mlsanity.com/paper/1704.02685/) (2017)
- [On the Properties of Neural Machine Translation: Encoder-Decoder Approaches](https://mlsanity.com/paper/1409.1259/) (2014)
- [Interpretable Explanations of Black Boxes by Meaningful Perturbation](https://mlsanity.com/paper/1704.03296/) (2017)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI](https://mlsanity.com/paper/1907.07374/) (2019)
- [RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism](https://mlsanity.com/paper/1608.05745/) (2016)
- [Understanding Deep Networks via Extremal Perturbations and Smooth Masks](https://mlsanity.com/paper/1910.08485/) (2019)
- [Explaining Time Series Predictions with Dynamic Masks](https://mlsanity.com/paper/2106.05303/) (2021)

## Built on by (most cited first, 23 of 23)

- [TimeX++: Learning Time-Series Explanations with Information Bottleneck](https://mlsanity.com/paper/2405.09308/) (2024)
- [Explaining Time Series via Contrastive and Locally Sparse Perturbations](https://mlsanity.com/paper/2401.08552/) (2024)
- [Deep Learning for Time Series Forecasting: A Survey](https://mlsanity.com/paper/2503.10198/) (2025)
- [TIMING: Temporality-Aware Integrated Gradients for Time Series Explanation](https://mlsanity.com/paper/2506.05035/) (2025)
- [FLEXtime: Filterbank learning to explain time series](https://mlsanity.com/paper/2411.05841/) (2024)
- [Time Interpret: a Unified Model Interpretability Library for Time Series](https://mlsanity.com/paper/2306.02968/) (2023)
- [FreqRISE: Explaining time series using frequency masking](https://mlsanity.com/paper/2406.13584/) (2024)
- [DeltaSHAP: Explaining Prediction Evolutions in Online Patient Monitoring with Shapley Values](https://mlsanity.com/paper/2507.02342/) (2025)
- [Dissecting Chronos: Sparse Autoencoders Reveal Causal Feature Hierarchies in Time Series Foundation Models](https://mlsanity.com/paper/2603.10071/) (2026)
- [Failure Modes of Time Series Interpretability Algorithms for Critical Care Applications and Potential Solutions](https://mlsanity.com/paper/2506.19035/) (2025)
- [xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truth](https://mlsanity.com/paper/2603.06781/) (2026)
- [Information Bottleneck Learning for Faithful Time Series Forecasting Explanations](https://mlsanity.com/paper/2607.28124/) (2026)
- [WinTSR: A Windowed Temporal Saliency Rescaling Method for Interpreting Time Series Deep Learning Models](https://mlsanity.com/paper/2412.04532/) (2024)
- [Delta-XAI: A Unified Framework for Explaining Prediction Changes in Online Time Series Monitoring](https://mlsanity.com/paper/2511.23036/) (2025)
- [When, How Long and How Much? Interpretable Neural Networks for Time Series Regression by Learning to Mask and Aggregate](https://mlsanity.com/paper/2512.03578/) (2025)
- [TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models](https://mlsanity.com/paper/2601.09776/) (2026)
- [Turning Time Series into Algebraic Equations: Symbolic Machine Learning for Interpretable Modeling of Chaotic Time Series](https://mlsanity.com/paper/2603.07261/) (2026)
- [Explainable AI Using Inherently Interpretable Components for Wearable-based Health Monitoring](https://mlsanity.com/paper/2603.12880/) (2026)
- [Global Explanations for Multivariate Time Series Forecasting Models via K-Order Markov Approximations](https://mlsanity.com/paper/2606.27599/) (2026)
- [Beyond Sufficiency: Time Series Explanation with Counterfactual Necessity](https://mlsanity.com/paper/2607.21573/) (2026)
- [Towards A Unified Information Bottleneck Framework for Time Series Explanations](https://mlsanity.com/paper/2608.25897/) (2026)
- [Explaining Time Series Forecasting with Horizon-Resolved Attribution](https://mlsanity.com/paper/2609.12639/) (2026)
- [Learnable Time-Frequency Masks for Explaining Time-Series Classifiers](https://mlsanity.com/paper/2609.29270/) (2026)

## 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.