Amortized Active Causal Induction with Deep Reinforcement Learning.
Yashas AnnadaniPanagiotis TigasStefan BauerAdam FosterPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- machine learning
- function approximation
- worst case
- optimal policy
- markov decision processes
- search tree
- running times
- deep learning
- subgroup discovery
- qualitative models
- reinforcement learning algorithms
- temporal difference
- model free
- state space
- learning process
- sequence prediction
- policy search
- program synthesis
- causal models
- function approximators
- search algorithm
- learning algorithm