Research map / Bandits and online learning / Online convex optimization and games
Online Meta-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 |
|---|---|---|---|
| Use FTML rather than traditional online-learning alternatives | Sequential vision tasks · Rainbow MNIST | task-learning efficiency | Helped |
| Use FTML rather than training independent models from scratch | Online learning across tasks · Five-Way CIFAR-100 | task-learning efficiency | Helped |
| Use FTML rather than training from scratch | Online learning across tasks · Sequential object pose prediction | task-learning efficiency | Helped |
| Use FTML rather than train on all previous data with task-specific fine-tuning | Online learning across tasks · Sequential object pose prediction | task performance | Helped |
| Adapt only the last layer instead of all layers (ablation) | FTML · Five-Way CIFAR-100 | task performance | Hurt |
| Use task-specific fine-tuning without explicitly meta-learning fine-tunability | FTL with fine-tuning · Sequential object pose prediction | task performance | Hurt |
Closest papers by meaning
- Online Fast Adaptation and Knowledge Accumulation (OSAKA): a New Approach to Continual Learning (NeurIPS 2020 · similarity 0.91)
- Online Constrained Meta-Learning: Provable Guarantees for Generalization (NeurIPS 2023 · similarity 0.90)
- Memory Efficient Online Meta Learning (ICML 2021 · similarity 0.90)
- Learning to Learn with Contrastive Meta-Objective (NeurIPS 2025 · similarity 0.87)
- Online Non-convex Learning in Dynamic Environments (NeurIPS 2024 · similarity 0.87)
- A Simple and Adaptive Learning Rate for FTRL in Online Learning with Minimax Regret of $\Theta(T^{2/3})$ and its Application to Best-of-Both-Worlds (NeurIPS 2024 · similarity 0.87)
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.