Research map / Time series / Causal and generative time series
Benchmarking Deep Learning Interpretability in Time Series Predictions
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply temporal saliency rescaling to gradient saliency | Grad on TCN · Middle Box | AUPR: 0.331 -> 0.399 | Helped |
| apply temporal saliency rescaling to gradient saliency | Grad on TCN · Moving Box | AUPR: 0.225 -> 0.335 | Helped |
| apply temporal saliency rescaling to gradient saliency | Grad on TCN · Moving Box | AUC accuracy drop: 95.35 -> 84.00 | Helped |
| apply TSR to SmoothGrad / Gradient SHAP | saliency maps · synthetic + MNIST time series | qualitative saliency quality | No clear effect |
| use standard saliency methods instead of random ranking | LSTM · synthetic time series benchmark | accuracy drop after masking | No clear effect |
| treat multivariate series as image with CNN instead of TCN | gradient saliency · Middle Box | qualitative saliency quality | Helped |
Closest papers by meaning
- Time series saliency maps: Explaining models across multiple domains (ICML 2026 · similarity 0.89)
- Improving Deep Learning Interpretability by Saliency Guided Training (NeurIPS 2021 · similarity 0.89)
- Input-Cell Attention Reduces Vanishing Saliency of Recurrent Neural Networks (NeurIPS 2019 · similarity 0.88)
- TIMESLIVER : SYMBOLIC-LINEAR DECOMPOSITION FOR EXPLAINABLE TIME SERIES CLASSIFICATION (ICLR 2026 · similarity 0.86)
- SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series (NeurIPS 2025 · similarity 0.86)
- DiffCAM: Data-Driven Saliency Maps by Capturing Feature Differences (CVPR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.