HSVI-based Online Minimax Strategies for Partially Observable Stochastic Games with Neural Perception Mechanisms.
Rui YanGabriel SantosGethin NormanDavid ParkerMarta KwiatkowskaPublished in: CoRR (2024)
Keyphrases
- partially observable stochastic games
- dynamic programming
- multi agent
- nash equilibrium
- neural network
- online learning
- partially observable markov decision processes
- worst case
- network architecture
- learning algorithm
- bayesian networks
- reinforcement learning
- cooperative
- domain independent
- online auctions
- neural mechanisms