Research map / Continual learning / Online and replay-based
Minimax Forward and Backward Learning of Evolving Tasks with Performance Guarantees
Veronica Alvarez, 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 LMRC instead of MER | continual learning · CLEAR | classification error: .10 -> .09 | Improved |
| use LMRC instead of Condor | concept drift adaptation · Yearbook | classification error: .14 -> .13 | Improved |
| use LMRC instead of MER | continual learning · Power Supply | classification error: .47 -> .30 | Improved |
| use LMRC instead of DriftSurf | concept drift adaptation · DomainNet | classification error: .32 -> .34 | Hurt |
| use LMRC instead of ELLA | lifelong learning · German | classification error: .33 -> .34 | Hurt |
| use LMRC instead of MER | continual learning · Usenet2 | classification error: .33 -> .33 | No clear effect |
| use LMRC instead of MER | continual learning · Covertype | classification error: .08 -> .08 | No clear effect |
| use LMRC instead of GEM | continual learning · Rotated MNIST | classification error: .36 -> .36 | No clear effect |
| forward and backward learning with k=100 tasks, n=10 samples vs single-task learning with n=100 | LMRC · Yearbook | classification error | Improved |
| add forward and backward learning over single-task learning | LMRC · Yearbook | classification error ratio | Improved |
| use LMRC instead of MER | running time · average per task | running time: 0.250 -> 0.275 | Hurt |
Closest papers by meaning
- Task-recency bias strikes back: Adapting covariances in Exemplar-Free Class Incremental Learning (NeurIPS 2024 · similarity 0.86)
- Prediction Error-based Classification for Class-Incremental Learning (ICLR 2024 · similarity 0.86)
- Mixture of Noise for Pre-Trained Model-Based Class-Incremental Learning (NeurIPS 2025 · similarity 0.85)
- Incremental Sequence Classification with Temporal Consistency (NeurIPS 2025 · similarity 0.85)
- Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers (ICML 2025 · similarity 0.85)
- Convergence and Implicit Bias of Gradient Descent on Continual Linear Classification (ICLR 2025 · 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.
