原书出处
Causal AI
current state is independent of the full history of states given the last state. Contrast this with the causal Markov property of causal DAGs: a node in the DAG is independent of indirect “ancestor” causes given its direct causal parents. We can see that when we view the MDP as a causal DAG, this Markovian assumption is equivalent to the causal Markov property. That means
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-12.xhtml
EPUB SHA-256:e55e7dc9d7966ea79311f2bc8d5bef7371a6dbb65240f11a2b3e44180b934e72