non-convex optimization

This new paper is a follow-up to our Anti-PGD paper (ICML 2022). This time we show that small perturbations induce explicit regularization, which we spell out for a few models. Many new research avenues open up from our Theorem 2.

I’m happy that our paper on AntiPGD, a version of stochastic gradient descent using anticorrelated perturbations, got accpeted to ICML 2022 in Baltimore, USA. You can read the PDF here. Looking forward to attend my first ML conference after Covid, present this paper, and have some scientific discussions.