Research map / Time series / Foundation models for time series
Temporal-Frequency Co-training for Time Series Semi-supervised Learning
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 |
|---|---|---|---|
| Co-train temporal and frequency views using each view's propagated pseudo-labels to guide the other | TS-TFC · 106 UCR datasets, 10% labeling | average accuracy: 0.754 -> 0.769 | Helped |
| Add warm-up training before using unlabeled data (ablation) | TS-TFC time-domain view · 106 UCR datasets, 10% labeling | average accuracy: 0.696 -> 0.769 | Helped |
| Add a queue that stores data from up to three consecutive mini-batches for label propagation (ablation) | TS-TFC time-domain view · 106 UCR datasets, 10% labeling | average accuracy: 0.744 -> 0.769 | Helped |
| Add temporal supervised contrastive learning (ablation) | TS-TFC time-domain view · 106 UCR datasets, 10% labeling | average accuracy: 0.745 -> 0.769 | Helped |
| Select high-quality propagated pseudo-labels using curriculum thresholds (ablation) | TS-TFC time-domain view · 106 UCR datasets, 10% labeling | average accuracy: 0.747 -> 0.769 | Helped |
| Add frequency supervised contrastive learning (ablation) | TS-TFC frequency-domain view · 106 UCR datasets, 10% labeling | average accuracy: 0.685 -> 0.769 | Helped |
| Select high-quality propagated pseudo-labels using curriculum thresholds (ablation) | TS-TFC frequency-domain view · 106 UCR datasets, 10% labeling | average accuracy: 0.687 -> 0.769 | Helped |
| Use both amplitude and phase rather than amplitude alone (ablation) | TS-F · 106 UCR datasets, 10% labeling | average accuracy: 0.682 -> 0.694 | Helped |
| Use both amplitude and phase rather than phase alone (ablation) | TS-F · 106 UCR datasets, 10% labeling | average accuracy: 0.581 -> 0.694 | Helped |
Closest papers by meaning
- Frequency-Masked Embedding Inference: A Non-Contrastive Approach for Time Series Representation Learning (AAAI 2025 · similarity 0.87)
- Boosting Transferability and Discriminability for Time Series Domain Adaptation (NeurIPS 2024 · similarity 0.86)
- TimesURL: Self-Supervised Contrastive Learning for Universal Time Series Representation Learning (AAAI 2024 · similarity 0.86)
- TimeSiam: A Pre-Training Framework for Siamese Time-Series Modeling (ICML 2024 · similarity 0.86)
- MF-CLR: Multi-Frequency Contrastive Learning Representation for Time Series (ICML 2024 · similarity 0.86)
- T-Rep: Representation Learning for Time Series using Time-Embeddings (ICLR 2024 · similarity 0.85)
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.