Reinforcing Language Agents via Policy Optimization with Action Decomposition.
Muning WenZiyu WanWeinan ZhangJun WangYing WenPublished in: CoRR (2024)
Keyphrases
- action selection
- agent receives
- robotic agents
- joint action
- dynamic logic
- multi agent
- multi agent systems
- action language
- optimization algorithm
- oriented programming
- cooperative
- action space
- expected reward
- autonomous agents
- intelligent agents
- mobile agents
- agent programming
- agent communication
- artificial agents
- reasoning about actions
- optimization problems
- programming language
- communicative acts
- natural language
- multiple agents
- agent learns
- reinforcement learning
- agent model
- dynamic environments
- agent architecture
- markov decision processes
- agent systems
- language learning
- agent oriented programming language
- decision making
- multi objective
- resource allocation
- software agents
- game theory
- incomplete information
- internal state
- decision process
- combinatorial auctions
- decomposition algorithm
- decision theoretic
- speech acts