原书出处
Causal AI
mechanics of how one might intervene in it is clear. Clarity here means you could run a real-world experiment that implemented the intervention, or, if the experiment were infeasible, unethical, or impossible, you could at least clearly articulate how the hypothetical experiment would work. “No causation without manipulation” is essentially trying to tether a causal model’s abstractions to experimental semantics.
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-7.xhtml
EPUB SHA-256:e55e7dc9d7966ea79311f2bc8d5bef7371a6dbb65240f11a2b3e44180b934e72