Research map / Bandits and online learning / Online convex optimization and games
Parameter-free, Dynamic, and Strongly-Adaptive Online 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 |
|---|---|---|---|
| Replace meta-algorithm aggregation of O(log T) interval learners with a residual learning reduction (second learner on extra dimension, added to first learner's outputs) | strongly-adaptive online linear optimization · theoretical regret analysis (OLO) | regret bound on any interval [a,b] | Helped |
| Generalize Cutkosky-Orabona constraint reduction to allow time-varying constraint sets and remove an extra factor of two from regret analysis | constrained online learning reduction · theoretical regret analysis (OLO) | regret bound and gradient norm of surrogate loss | Helped |
| Decompose strongly-adaptive regret bound into intervals with bounded path-length to obtain optimal dependence on path-length P (sqrt(P)) in dynamic regret | dynamic regret of the strongly-adaptive algorithm · theoretical regret analysis (OLO) | dynamic regret bound dependence on path length P | Helped |
| Per-update runtime of the proposed algorithm compared to prior second-order strongly-adaptive method | strongly-adaptive second-order algorithms · theoretical runtime | runtime per update | Helped |
Closest papers by meaning
- Non-stationary Projection-Free Online Learning with Dynamic and Adaptive Regret Guarantees (AAAI 2024 · similarity 0.92)
- Non-stationary Online Learning for Curved Losses: Improved Dynamic Regret via Mixability (ICML 2025 · similarity 0.90)
- From Average Sensitivity to Small-Loss Regret Bounds under Random-Order Model (NeurIPS 2026 · similarity 0.90)
- 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 0.90)
- A Simple and Optimal Approach for Universal Online Learning with Gradient Variations (NeurIPS 2024 · similarity 0.90)
- Adaptive Regret for Bandits Made Possible: Two Queries Suffice (ICLR 2024 · similarity 0.90)
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.