causal

Marry Learning With Error-Tolerant Reasoning

Combining classical logic with machine learning fails because logic is deterministic and unforgiving of errors and inconsistencies while learning is forgiving of them; the right approach is to distort reasoning to be compatible with learning, so that 'A implies B and B implies C therefore A implies C' holds only with high probability, enabling more reliable, authoritative AI systems.

causalpending

Speaker

Leslie Valiant

Evidence Quote

“the approach I've been trying to pursue is to impose on, is to marry learning and reasoning, but distort the reasoning process to be compatible with learning. So you need some reasoning, which forgives errors and things are... Correct under a certain probability. If A implies B and B implies C, then A implies C would only be true with high probability.”

Source

272 | Leslie Valiant on Learning and Educability in Computers and People— Sean Carroll's Mindscape
Created: 6/13/2026, 12:22:31 AM

Your notes

Loading notes...