Research map / Time series / Foundation models for time series
Adversarial Dynamic Shapelet Networks
Qianli Ma, Wanqing Zhuang, Sen Li, Desen Huang, Garrison Cottrell
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add adversarial loss and diversity regularization (ablation) | ADSN · 18 UCR + 8 UEA datasets | average rank: 4.3 -> 2.0 | Improved |
| add adversarial loss (ablation) | ADSN (with diversity term) · 18 UCR + 8 UEA datasets | average rank: 3.7 -> 2.0 | Improved |
| add diversity regularization (ablation) | ADSN (with adversarial loss) · 18 UCR + 8 UEA datasets | average rank: 3.1 -> 2.0 | Improved |
| generate sample-specific shapelets dynamically instead of fixed learned shapelets (LTS) | shapelet learning · 18 UCR + 8 UEA datasets | average rank: 5.0 -> 2.0 | Improved |
| increase lambda_adv (adversarial weight) (ablation) | ADSN · shapelet visualization | similarity to real subsequences | Improved |
| increase lambda_div from 0 (ablation) | ADSN · shapelet visualization | shapelet diversity | Improved |
| ADSN versus FCN and ResNet | ADSN · 18 UCR + 8 UEA datasets | average rank: 3.4 vs 3.5 (FCN) | No clear effect |
Closest papers by meaning
- Learning Soft Sparse Shapes for Efficient Time-Series Classification (ICML 2025 · similarity 0.87)
- ShapeX: Shapelet-Driven Post Hoc Explanations for Time Series Classification Models (NeurIPS 2025 · similarity 0.85)
- Diffusion Language-Shapelets for Semi-supervised Time-Series Classification (AAAI 2024 · similarity 0.84)
- TVNet: A Novel Time Series Analysis Method Based on Dynamic Convolution and 3D-Variation (ICLR 2025 · similarity 0.83)
- Time Series Representations with Hard-Coded Invariances (ICML 2025 · similarity 0.83)
- Abstracted Shapes as Tokens - A Generalizable and Interpretable Model for Time-series Classification (NeurIPS 2024 · similarity 0.82)
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.
