Research map / Bandits and online learning / Multi-armed bandits
Adaptive Algorithms for Relaxed Pareto Set Identification
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 PSI-Unif-Elim by APE (k=K) | PSI algorithm · Random Bernoulli, K=5, D=2 | average sample complexity: 190625 -> 150844 | Helped |
| replace PSI-Unif-Elim by APE (k=K) | PSI algorithm · Random Bernoulli, K=5, D=4 | average sample complexity: 157584 -> 124001 | Helped |
| replace PSI-Unif-Elim by APE (k=K) | PSI algorithm · Random Bernoulli, K=5, D=8 | average sample complexity: 27270 -> 20336 | Helped |
| relax to identify at most k=1 Pareto arm instead of all (ablation) | APE · Random Bernoulli, K=5, D=2 | average sample complexity: 150844 -> 811 | Helped |
| relax to at most k=2 arms instead of k=5 (ablation) | APE · Random Bernoulli, K=5, D=4 | average sample complexity: 124001 -> 6410 | Helped |
| relax to at most k=3 arms instead of k=5 (ablation) | APE · Random Bernoulli, K=5, D=8 | average sample complexity: 20336 -> 405 | Helped |
| increase k from 2 to 3 (ablation) | APE · Random Bernoulli, K=5, D=2 | average sample complexity: 39530 -> 109020 | Hurt |
| increase dimension D from 2 to 8 for APE-5 | APE · Random Bernoulli, K=5 | average sample complexity: 150844 -> 20336 | Helped |
| use k>3 relaxation vs unrelaxed exact PSI | APE · COV-BOOST real-world scenario | empirical sample complexity | Helped |
| use k=1 or k=2 relaxation (ablation) | APE · COV-BOOST real-world scenario | empirical sample complexity | Helped |
| add eps2 relaxation (eps2 from 0.01 to 0.05) (ablation) | APE · Random Bernoulli, K=5, D=2 | average sample complexity: 86% | Helped |
Closest papers by meaning
- Constrained Pareto Set Identification with Bandit Feedback (ICML 2025 · similarity 0.88)
- Thompson Sampling for Multi-Objective Linear Contextual Bandit (NeurIPS 2025 · similarity 0.84)
- Noise-Adaptive Confidence Sets for Linear Bandits and Application to Bayesian Optimization (ICML 2024 · similarity 0.84)
- Hierarchize Pareto Dominance in Multi-Objective Stochastic Linear Bandits (AAAI 2024 · similarity 0.84)
- Preference Optimization on Pareto Sets: On a Theory of Multi-Objective Optimization (NeurIPS 2025 · similarity 0.83)
- On the Asymptotic Optimality of Confidence Interval Based Algorithms for Fixed Confidence MABs (AAAI 2025 · similarity 0.83)
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.