Understanding the Convergence of Optimization Algorithms for Minimax Machine Learning

<<< All talks

Understanding the Convergence of Optimization Algorithms for Minimax Machine Learning

When	Friday, January 14, 2022, 3:00 PM – 4:00 PM (MT)
Where	WEB 1250

Abstract

The past decade has witnessed the great success of deep learning in broad societal and commercial applications. However, conventional deep learning relies on wildly fitting data with neural networks, which is known to produce models that lack resilience. For instance, models used in facial recognition and healthcare are known to be biased toward people of a certain race or gender. Models used in autonomous driving are vulnerable to malicious attacks, i.e., putting an art sticker on a stop sign may force the model to classify it as a speed limit sign. Therefore, the next-generation deep learning paradigm aims to deliver resilient models that promote robustness to malicious attacks, fairness among users, and privacy preservation, and this can be realized by leveraging the emerging minimax machine learning framework. In this talk, I will present three gradient-descent-ascent (GDA) type of optimization algorithms for solving different classes of nonconvex minimax machine learning problems. Then, I will present a principled nonconvex minimax optimization theory that establishes the global convergence and convergence rates of these algorithms.

Speaker

Tags: algorithms & theory deep learning machine learning optimization

---

Part of the Data Science Seminar. Something wrong on this page? Edit _data/talks/2022-01-14-yi-zhou.toml.