A nonlinear causal model reduction framework learns interpretable high-level causes of reward in RL policies, with uniqueness guarantees for a class of additive noise models.
Mastering the game of Go with deep neural networks and tree search
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Learning Nonlinear Causal Reductions to Explain Reinforcement Learning Policies
A nonlinear causal model reduction framework learns interpretable high-level causes of reward in RL policies, with uniqueness guarantees for a class of additive noise models.