3 ms·
Feynman Symbolic Regression Database
- itsthecourier 2y agoThat's a really hard problem, I wonder how stuff like topology and combinatorial mathematics are related to this, do they apply to this kind of problems?
- bionhoward 2y agoYes. Topology of the function ASTs and combinatorics of the various ways to assign primitives (fn) and terminals (let) to the nodes of the function. The answers to the challenge are words so obey a zipfian distribution
- steve_gh 2y agoThere was a piece here on HN about a week ago on Kolmagorov Arnold network (KANs) as an alternative to NNs. https://news.ycombinator.com/item?id=40219205 https://news.ycombinator.com/item?id=40219205 If I understood it correctly, a KAN would be a suitable network to work on this problem, as it's activation functions are in the links (and can be any function).
- thesz 2y agoPage 15 of KAN paper deals with Feynman datasets.
- wave_1 2y agoIf we can figure out symbolic regression in a universal fashion that would be AGI in a way, right?
- Nevermark 2y agoYes, I think so - especially if it was fast. A very fast at scale universal symbolic regression machine would be formidable.