# TSInsight: A local-global attribution framework for interpretability in time-series data

arXiv: https://arxiv.org/abs/2004.02958 · HTML: https://mlsanity.com/paper/2004.02958/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Apr 2020.

## Built on by (most cited first, 8 of 8)

- [Explainable Artificial Intelligence (XAI) on TimeSeries Data: A Survey](https://mlsanity.com/paper/2104.00950/) (2021)
- [Time Series Model Attribution Visualizations as Explanations](https://mlsanity.com/paper/2109.12935/) (2021)
- [TSInterpret: A unified framework for time series interpretability](https://mlsanity.com/paper/2208.05280/) (2022)
- [Anomaly Attribution of Multivariate Time Series using Counterfactual Reasoning](https://mlsanity.com/paper/2109.06562/) (2021)
- [XTSC-Bench: Quantitative Benchmarking for Explainers on Time Series Classification](https://mlsanity.com/paper/2310.14957/) (2023)
- [PatchX: Explaining Deep Models by Intelligible Pattern Patches for Time-series Classification](https://mlsanity.com/paper/2102.05917/) (2021)
- [Explainable Deep Learning Framework for Human Activity Recognition](https://mlsanity.com/paper/2408.11552/) (2024)
- [SSET: Swapping-Sliding Explanation for Time Series Classifiers in Affect Detection](https://mlsanity.com/paper/2410.12996/) (2024)

## Abstract (arXiv)

With the rise in the employment of deep learning methods in safety-critical scenarios, interpretability is more essential than ever before. Although many different directions regarding interpretability have been explored for visual modalities, time-series data has been neglected with only a handful of methods tested due to their poor intelligibility. We approach the problem of interpretability in a novel way by proposing TSInsight where we attach an auto-encoder to the classifier with a sparsity-inducing norm on its output and fine-tune it based on the gradients from the classifier and a reconstruction penalty. TSInsight learns to preserve features that are important for prediction by the classifier and suppresses those that are irrelevant i.e. serves as a feature attribution method to boost interpretability. In contrast to most other attribution frameworks, TSInsight is capable of generating both instance-based and model-based explanations. We evaluated TSInsight along with 9 other commonly used attribution methods on 8 different time-series datasets to validate its efficacy. Evaluation results show that TSInsight naturally achieves output space contraction, therefore, is an effective tool for the interpretability of deep time-series models.