Abstract:We present a novel gradient-free algorithm to solve a convex stochastic optimization problem, such as those encountered in medicine, physics, and machine learning (e.g., adversarial multi-armed bandit problem), where the objective function can only be computed through numerical simulation, either as the result of a real experiment or as feedback given by the function evaluations from an adversary. Thus we suppose that only a black-box access to the function values of the objective is available, possibly corrupted by adversarial noise: deterministic or stochastic. The noisy setup can arise naturally from modeling randomness within a simulation or by computer discretization, or when exact values of function are forbidden due to privacy issues, or when solving non-convex problems as convex ones with an inexact function oracle. By exploiting higher-order smoothness, fulfilled, e.g., in logistic regression, we improve the performance of zero-order methods developed under the assumption of classical smoothness (or having a Lipschitz gradient). The proposed algorithm enjoys optimal oracle complexity and is designed under an overparameterization setup, i.e., when the number of model parameters is much larger than the size of the training dataset. Overparametrized models fit to the training data perfectly while also having good generalization and outperforming underparameterized models on unseen data. We provide convergence guarantees for the proposed algorithm under both types of noise. Moreover, we estimate the maximum permissible adversarial noise level that maintains the desired accuracy in the Euclidean setup, and then we extend our results to a non-Euclidean setup. Our theoretical results are verified on the logistic regression problem.

Stochastic Adversarial Noise in the "Black Box" Optimization Problem

Stochastic Optimization with Non-stationary Noise

High Probability Complexity Bounds for Non-Smooth Stochastic Optimization with Heavy-Tailed Noise

High-Probability Complexity Bounds for Non-smooth Stochastic Convex Optimization with Heavy-Tailed Noise

Improved Iteration Complexity in Black-Box Optimization Problems under Higher Order Smoothness Function Condition

The "Black-Box" Optimization Problem: Zero-Order Accelerated Stochastic Method via Kernel Approximation

The Black-Box Optimization Problem: Zero-Order Accelerated Stochastic Method via Kernel Approximation

Stochastic Nonsmooth Convex Optimization with Heavy-Tailed Noises: High-Probability Bound, In-Expectation Rate and Initial Distance Adaptation

On Convergence of Adam for Stochastic Optimization under Relaxed Assumptions

Breaking the Heavy-Tailed Noise Barrier in Stochastic Optimization Problems

Stochastic Optimization with Non-stationary Noise: the Power of Moment Estimation

May the Noise be with you: Adversarial Training without Adversarial Examples

Accelerated zero-order SGD under high-order smoothness and overparameterized regime

First Order Stochastic Optimization with Oblivious Noise

Nonlinear Stochastic Gradient Descent and Heavy-tailed Noise: A Unified Framework and High-probability Guarantees

Byzantine-robust decentralized stochastic optimization with stochastic gradient noise-independent learning error

Adversarial Online Learning with noise

Online Learning: Stochastic and Constrained Adversaries

Stochastic Approximation for Expectation Objective and Expectation Inequality-Constrained Nonconvex Optimization

Algorithms with Gradient Clipping for Stochastic Optimization with Heavy-Tailed Noise

A Stochastic Objective-Function-Free Adaptive Regularization Method with Optimal Complexity