Unregularized limit of stochastic gradient method for Wasserstein distributionally robust optimization
By: Tam Le
Potential Business Impact:
Makes computer learning better with uncertain information.
Distributionally robust optimization offers a compelling framework for model fitting in machine learning, as it systematically accounts for data uncertainty. Focusing on Wasserstein distributionally robust optimization, we investigate the regularized problem where entropic smoothing yields a sampling-based approximation of the original objective. We establish the convergence of the approximate gradient over a compact set, leading to the concentration of the regularized problem critical points onto the original problem critical set as regularization diminishes and the number of approximation samples increases. Finally, we deduce convergence guarantees for a projected stochastic gradient method. Our analysis covers a general machine learning situation with an unbounded sample space and mixed continuous-discrete data.
Similar Papers
Wasserstein Distributionally Robust Nonparametric Regression
Machine Learning (Stat)
Makes computer predictions better even with bad data.
Optimal Regularization Under Uncertainty: Distributional Robustness and Convexity Constraints
Optimization and Control
Makes computer guesses good even with bad information.
Provable Robust Overfitting Mitigation in Wasserstein Distributionally Robust Optimization
Machine Learning (CS)
Makes AI smarter and less fooled by tricks.