Research map / Bandits and online learning / Online convex optimization and games
Mixing Predictions for Online Metric Algorithms
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 |
|---|---|---|---|
| Combine ℓ predictors with full access and compare against their unrestricted dynamic combination | MTS · MTS, full access, DYN | competitive ratio: O(ℓ^2); lower bound Ω(ℓ^2) | Mixed |
| Constrain the benchmark to at most m predictor switches | Full-access predictor combination · MTS, full access, DYN^{≤m} | competitive ratio: (1+ε)^2 | Helped |
| Allow access to only one queried predictor at each time step | Bandit predictor combination · MTS, bandit access, DYN^{≤m} | competitive ratio: (1+\epsilon)^3 | Mixed |
| Use the server-named-by-a-predictor benchmark \widetilde{DYN} instead of the configuration benchmark DYN | k-server · k-server on the line | competitive ratio: k; Ω(log k) | Hurt |
Closest papers by meaning
- Online metric algorithms with untrusted predictions (ICML 2020 · similarity 0.90)
- Learning-Augmented Algorithms for MTS with Bandit Access to Multiple Predictors (ICML 2025 · similarity 0.88)
- Learning-Augmented Algorithms with Explicit Predictors (NeurIPS 2024 · similarity 0.86)
- A General Online Algorithm for Optimizing Complex Performance Metrics (ICML 2024 · similarity 0.86)
- Barely Random Algorithms and Collective Metrical Task Systems (NeurIPS 2024 · similarity 0.85)
- Augmenting Online Algorithms for Knapsack Problem with Total Weight Information (AAAI 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.