Implicit Regularization in Matrix Factorization
–Neural Information Processing Systems
We study implicit regularization when optimizing an underdetermined quadratic objective over a matrix X with gradient descent on a factorization of X. We conjecture and provide empirical and theoretical evidence that with small enough step sizes and initialization close enough to the origin, gradient descent on a full dimensional factorization converges to the minimum nuclear norm solution.
Neural Information Processing Systems
May-26-2025, 13:53:00 GMT
- Technology: