jesustabares

3 ms·

1. 1.▲
   Laimark – 8B LLM that self-improves. Consumer GPU(github.com)

   3 points by jesustabares 6mo ago|0 comments
2. 2.▲
   by jesustabares 6mo ago

   This article inspired us to test whether EML trees can be trained using gradient descent for symbolic regression: instead of searching for the right tree, is it possible to optimize one from start to finish? We implemented the EML trees as