SoK: Adversarial Machine Learning Attacks and Defences in Multi-Agent Reinforcement Learning

TL;DR

This paper surveys adversarial machine learning attacks and defenses in multi-agent reinforcement learning, proposing new frameworks and perspectives to understand attack methods and identify research gaps.

Contribution

It introduces novel frameworks for modeling AML attacks on MARL and a new perspective using Attack Vectors, advancing understanding of attack mechanisms.

Findings

Identified key attack vectors in AML for MARL

Developed frameworks addressing attack means and tempo

Highlighted knowledge gaps and future research directions

Abstract

Multi-Agent Reinforcement Learning (MARL) is vulnerable to Adversarial Machine Learning (AML) attacks and needs adequate defences before it can be used in real world applications. We have conducted a survey into the use of execution-time AML attacks against MARL and the defences against those attacks. We surveyed related work in the application of AML in Deep Reinforcement Learning (DRL) and Multi-Agent Learning (MAL) to inform our analysis of AML for MARL. We propose a novel perspective to understand the manner of perpetrating an AML attack, by defining Attack Vectors. We develop two new frameworks to address a gap in current modelling frameworks, focusing on the means and tempo of an AML attack against MARL, and identify knowledge gaps and future avenues of research.