Causal reinforcement learning, organized around Bareinboim's 9-task taxonomy — confounded offline→online RL, POMIS, counterfactual policies, transportability, discovery, imitation, curricula, shaping, games. Honest about scope. pip install causalrl
python reinforcement-learning pytorch sensitivity-analysis causal-inference gymnasium bandits counterfactual do-calculus causal-discovery off-policy-evaluation causal-reinforcement-learning causal-rl structural-causal-models transportability causal-bandits
-
Updated
Jun 9, 2026 - Python