PI-CMDP combines causal backdoor identification under LOA, Markov state compression, and physics-guided doubly-robust estimation to achieve higher constraint repair success rates with fewer samples than baselines on the TPS benchmark.
Regret analysis of causal reward maximization.NeurIPS
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Physics-Informed Causal MDPs for Sequential Constraint Repair in Engineering Simulation Pipelines
PI-CMDP combines causal backdoor identification under LOA, Markov state compression, and physics-guided doubly-robust estimation to achieve higher constraint repair success rates with fewer samples than baselines on the TPS benchmark.