Dynamic ensemble selection based on Deep Neural Network Uncertainty Estimation for Adversarial Robustness

TL;DR

This paper proposes a dynamic ensemble selection method based on neural network uncertainty estimation to enhance adversarial robustness without sacrificing accuracy.

Contribution

It introduces a model-level dynamic ensemble approach using Dirichlet distribution and diversity constraints to improve defense against white-box attacks.

Findings

Significant robustness improvements over static models.

Maintains accuracy while increasing adversarial resistance.

Effective dynamic selection based on uncertainty values.

Abstract

The deep neural network has attained significant efficiency in image recognition. However, it has vulnerable recognition robustness under extensive data uncertainty in practical applications. The uncertainty is attributed to the inevitable ambient noise and, more importantly, the possible adversarial attack. Dynamic methods can effectively improve the defense initiative in the arms race of attack and defense of adversarial examples. Different from the previous dynamic method depend on input or decision, this work explore the dynamic attributes in model level through dynamic ensemble selection technology to further protect the model from white-box attacks and improve the robustness. Specifically, in training phase the Dirichlet distribution is apply as prior of sub-models' predictive distribution, and the diversity constraint in parameter space is introduced under the lightweight sub-models to construct alternative ensembel model spaces. In test phase, the certain sub-models are dynamically selected based on their rank of uncertainty value for the final prediction to ensure the majority accurate principle in ensemble robustness and accuracy. Compared with the previous dynamic method and staic adversarial traning model, the presented approach can achieve significant robustness results without damaging accuracy by combining dynamics and diversity property.