Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning.
Natasha JaquesAngeliki LazaridouEdward HughesÇaglar GülçehrePedro A. OrtegaDJ StrouseJoel Z. LeiboNando de FreitasPublished in: ICML (2019)
Keyphrases
- social influence
- intrinsic motivation
- reinforcement learning
- multi agent
- social interaction
- social networks
- statistically significant
- user preferences
- college students
- cooperative
- online communities
- multi agent systems
- network structure
- job satisfaction
- learning process
- learning algorithm
- learning agents
- cognitive style
- intelligent tutoring systems
- multiple agents
- e learning