Research map / Bandits and online learning / Online convex optimization and games
The Primal-Dual method for Learning Augmented 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 |
|---|---|---|---|
| Adjust primal update rates toward sets in the predicted cover | PDLA for online weighted set cover · Online weighted set cover | cost relative to predicted solution: O(1/(1-λ)) | Helped |
| Use predicted vacation length to choose aggressive or cautious buying updates | PDLA for ski rental · Ski rental | competitive ratio to OPT: 1/(1-e(-λ)) | Helped |
| Use predicted Bahncard purchase times to choose aggressive or cautious updates | PDLA for Bahncard · Bahncard | competitive ratio to OPT: (e^λ-β)/(e^λ-1) | Helped |
| Use predicted acknowledgement times to guide update rates | PDLA for dynamic TCP acknowledgement · Dynamic TCP acknowledgement | competitive ratio to OPT: 1/(1-e^{-λ}) | Helped |
| Lower λ when the prediction is perfect | PDLA for dynamic TCP acknowledgement · Dynamic TCP acknowledgement | competitive ratio | Helped |
| Increase prediction replacement noise | PDLA for dynamic TCP acknowledgement · Poisson, Pareto, and iterated Poisson packet arrivals | competitive ratio | Mixed |
Closest papers by meaning
- Learning-Augmented Online Minimization with Dual Predictions (ICML 2026 · similarity 0.90)
- Learning Augmented Exact Exponential Algorithms (NeurIPS 2026 · similarity 0.87)
- Learning-Augmented Algorithms with Explicit Predictors (NeurIPS 2024 · similarity 0.87)
- Online Learning under Budget and ROI Constraints via Weak Adaptivity (ICML 2024 · similarity 0.85)
- Overcoming Brittleness in Pareto-Optimal Learning Augmented Algorithms (NeurIPS 2024 · similarity 0.85)
- On the Universal Near Optimality of Hedge in Combinatorial Settings (NeurIPS 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.