Learning sudoku by doing gradient descent on a linear program(mxkopy.github.io)
mxkopy.github.io
Learning sudoku by doing gradient descent on a linear program
https://mxkopy.github.io/learning-sudoku-constraints/
1 comments
In this post I go over training LPs using cvxpylayers, but I think the more interesting discussion lies at the end, where I argue that current AI models lack the ability to reason counterfactually in discrete settings & that LPs can bridge this gap.