Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2406.13584 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910736861626368 |
|---|---|
| author | Brüsch, Thea Wickstrøm, Kristoffer Knutsen Schmidt, Mikkel N. Alstrøm, Tommy Sonne Jenssen, Robert |
| author_facet | Brüsch, Thea Wickstrøm, Kristoffer Knutsen Schmidt, Mikkel N. Alstrøm, Tommy Sonne Jenssen, Robert |
| contents | Time-series data are fundamentally important for many critical domains such as healthcare, finance, and climate, where explainable models are necessary for safe automated decision making. To develop explainable artificial intelligence in these domains therefore implies explaining salient information in the time series. Current methods for obtaining saliency maps assume localized information in the raw input space. In this paper, we argue that the salient information of a number of time series is more likely to be localized in the frequency domain. We propose FreqRISE, which uses masking-based methods to produce explanations in the frequency and time-frequency domain, and outperforms strong baselines across a number of tasks. The source code is available here: \url{https://github.com/theabrusch/FreqRISE}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_13584 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FreqRISE: Explaining time series using frequency masking Brüsch, Thea Wickstrøm, Kristoffer Knutsen Schmidt, Mikkel N. Alstrøm, Tommy Sonne Jenssen, Robert Machine Learning Time-series data are fundamentally important for many critical domains such as healthcare, finance, and climate, where explainable models are necessary for safe automated decision making. To develop explainable artificial intelligence in these domains therefore implies explaining salient information in the time series. Current methods for obtaining saliency maps assume localized information in the raw input space. In this paper, we argue that the salient information of a number of time series is more likely to be localized in the frequency domain. We propose FreqRISE, which uses masking-based methods to produce explanations in the frequency and time-frequency domain, and outperforms strong baselines across a number of tasks. The source code is available here: \url{https://github.com/theabrusch/FreqRISE}. |
| title | FreqRISE: Explaining time series using frequency masking |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2406.13584 |