Improving Model Robustness by Adaptively Correcting Perturbation Levels with Active Queries

TL;DR

This paper introduces an adaptive method for improving machine learning model robustness by interactively querying optimal perturbation levels from humans, rather than using fixed perturbations, leading to better robustness with minimal queries.

Contribution

It proposes a novel active learning framework that adaptively adjusts perturbation levels for training examples, enhancing robustness efficiently.

Findings

Significant robustness improvement with few human queries.

Theoretical validation of the adaptive perturbation approach.

Effective sampling strategy for querying perturbation levels.

Abstract

In addition to high accuracy, robustness is becoming increasingly important for machine learning models in various applications. Recently, much research has been devoted to improving the model robustness by training with noise perturbations. Most existing studies assume a fixed perturbation level for all training examples, which however hardly holds in real tasks. In fact, excessive perturbations may destroy the discriminative content of an example, while deficient perturbations may fail to provide helpful information for improving the robustness. Motivated by this observation, we propose to adaptively adjust the perturbation levels for each example in the training process. Specifically, a novel active learning framework is proposed to allow the model to interactively query the correct perturbation level from human experts. By designing a cost-effective sampling strategy along with a new query type, the robustness can be significantly improved with a few queries. Both theoretical analysis and experimental studies validate the effectiveness of the proposed approach.