Research map / Continual learning / Online and replay-based
Supervised Learning with Evolving Tasks and Performance Guarantees
Verónica Álvarez, Santiago Mazuelas, Jose A. Lozano
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 |
|---|---|---|---|
| use smoothing-based forward learning instead of sliding windows (W=5, 25, 45) for evolving tasks | MRC with evolving-task mean estimation · Numerical ESS comparison, j=50 | effective sample size | Improved |
| add backward learning from succeeding tasks to forward learning (ablation) | MRC with evolving tasks · Numerical ESS comparison, k=100 tasks, j=50 | effective sample size | Improved |
| use sliding windows with small versus large window size (ablation) | sliding-window baseline · Numerical ESS comparison | effective sample size | Mixed |
| use independent-changes (random-walk) task assumption instead of i.i.d. tasks | task distribution model · Airlines and UTKFaces partial autocorrelation | partial autocorrelation at lag 1 | Improved |
Closest papers by meaning
- Minimax Forward and Backward Learning of Evolving Tasks with Performance Guarantees (NeurIPS 2023 · similarity 0.88)
- Optimal Task Order for Continual Learning of Multiple Tasks (ICML 2025 · similarity 0.85)
- CLDyB: Towards Dynamic Benchmarking for Continual Learning with Pre-trained Models (ICLR 2025 · similarity 0.84)
- Incremental Sequence Classification with Temporal Consistency (NeurIPS 2025 · similarity 0.84)
- Adaptive Retention & Correction: Test-Time Training for Continual Learning (ICLR 2025 · similarity 0.84)
- Continual Learning with Global Alignment (NeurIPS 2024 · 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.
