Improving Multimodal Interactive Agents with Reinforcement Learning from Human Feedback.
Josh AbramsonArun AhujaFederico CarnevalePetko GeorgievAlex GoldinAlden HungJessica LandonJirka LhotkaTimothy O. LillicrapAlistair MuldalGeorge PowellAdam SantoroGuy ScullySanjana SrivastavaTamara von GlehnGreg WayneNathaniel WongChen YanRui ZhuPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- multi agent
- artificial agents
- multi agent systems
- virtual characters
- learning agents
- multiagent systems
- multiple agents
- action selection
- multi agent reinforcement learning
- cooperative
- intelligent agents
- single agent
- multiagent learning
- function approximation
- software agents
- reinforcement learning algorithms
- agent teams
- multi modal
- reward signal
- multi agent environments
- decision making
- agent interactions
- learning algorithm
- learning agent
- machine learning
- conversational agents
- autonomous agents
- reinforcement learning agents
- virtual environment
- partial observability
- personality types
- agent receives
- human operators
- multi party
- agent architecture
- markov decision processes
- user interaction
- relevance feedback
- state space
- virtual agents
- personality traits
- model free
- human communication
- evolutionary learning
- resource allocation
- computer graphics
- electronic commerce
- dynamic programming