Causal Reasoning from Meta-reinforcement Learning.
Ishita DasguptaJane X. WangSilvia ChiappaJovana MitrovicPedro A. OrtegaDavid RaposoEdward HughesPeter W. BattagliaMatthew BotvinickZeb Kurth-NelsonPublished in: CoRR (2019)
Keyphrases
- causal reasoning
- reinforcement learning
- causal models
- function approximation
- machine learning
- state space
- multi agent
- knowledge representation
- robotic control
- model free
- bayes net
- reinforcement learning algorithms
- meta level
- directed acyclic graph
- learning algorithm
- learning problems
- temporal difference
- optimal policy
- search results clustering
- reinforcement learning methods
- default logic
- relational databases