Re:AlphaGeometry: An Olympiad-level AI system for geometry
by Imnimo on Wednesday January 17, 2024 @02:47PM
(#39032487)
>A transformer is trained on millions of elementary geometry theorems and used as brute search for a proof, which because of the elementary geometry context has both a necessarily elementary structure and can be easily symbolically judged as true or false. When the brute search fails, an extra geometric construction is randomly added (like adding a midpoint of a certain line) to see if brute search using that extra raw material might work.
I don't think this is quite right. The brute force search is performed by a symbolic solver, not the transformer. When it runs out of new deductions, the transformer is asked to suggest possible extra constructions (not randomly added).
I don't think this is quite right. The brute force search is performed by a symbolic solver, not the transformer. When it runs out of new deductions, the transformer is asked to suggest possible extra constructions (not randomly added).