Research map / Learning theory and algorithms / Submodular and game algorithms
Incremental Multiple Oracle
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Incremental Multiple Oracle, a modification of the multiple oracle algorithm that computes approximate mixed Nash equilibria of continuous-action games while keeping fixed-size pure strategy sets per player.
Problem
Computing approximate mixed-strategy Nash equilibria of continuous-action games, where prior double/multiple oracle methods grow strategy sets and thus require unbounded memory.
Approach
Modifies the multiple oracle algorithm (an extension of double oracle to multiple players and continuous actions) so that each player's pure strategy set has fixed cardinality, requiring only constant memory.
Key results
- Maintains fixed-cardinality pure strategy sets per player, so only a constant amount of memory is needed, unlike prior methods that grow strategy sets.
Limitations
- Full text unavailable; experimental details and convergence guarantees could not be verified from the abstract.
Experimental setup
Baselines: multiple oracle double oracle
Builds on: double oracle multiple oracle
Closest papers by meaning
- Double Oracle Algorithm for Computing Equilibria in Continuous Games (AAAI 2021 · similarity 0.92)
- Virtual Double Oracle: Faster Convergence by Leveraging Strategies That Were Watching All Along (NeurIPS 2026 · similarity 0.90)
- Exclusion Method for Finding Nash Equilibrium in Multiplayer Games (AAAI 2017 · similarity 0.89)
- Tractable Algorithms for Approximate Nash Equilibria in Generalized Graphical Games with Tree Structure (AAAI 2017 · similarity 0.89)
- Can Reinforcement Learning Solve Asymmetric Combinatorial-Continuous Zero-Sum Games? (ICLR 2025 · similarity 0.87)
- Commitment to Sparse Strategies in Two-Player Games (AAAI 2025 · similarity 0.86)
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.