DEFENDING STRATEGIES AGAINST ADVERSARIAL ATTACKS IN RETRIEVAL SYSTEMS

Abstract

During history, retrieval systems become more complicated in their architecture design and work principles. The system that gathers text and visual data from the internet must classify the data and store it as the set of metadata. The modern AI classifiers that are used in retrieval systems might be tricked by skilled intruders who use adversarial attacks on the retrieval system. The goal of this paper is to review different strategies of attacks and defenses, describe state-of-the-art methods from both sides, and show how important the development of HPC is in protecting systems.