Adversarial Robustness of Deep Reinforcement Learning Based Dynamic Recommender Systems.

ABSTRACT: Adversarial attacks, e.g., adversarial perturbations of the input and adversarial samples, pose significant challenges to machine learning and deep learning techniques, including interactive recommendation systems. The latent embedding space of those techniques makes adversarial attacks challenging to detect at an early stage. Recent advance in causality shows that counterfactual can also be considered one of the ways to generate the adversarial samples drawn from different distribution as the training samples. We propose to explore adversarial examples and attack agnostic detection on reinforcement learning (RL)-based interactive recommendation systems. We first craft different types of adversarial examples by adding perturbations to the input and intervening on the casual factors. Then,

SUBMITTER: Wang S 

PROVIDER: S-EPMC9110778 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature