Increasing Trust in Development Processes Using Robust, Data-Driven Markov Games

In this article, we consider how to increase trust in development processes in which there is a risk for adversarial manipulation, and the adversary's objectives and resources are either ill-specified, imprecisely specified, or unknown. In such problems, we must hedge against the risk of misapprehension of attacker objectives and resources, which is further complicated in the absence of adversarial training data. We show how to model dynamic agent interaction, on the basis of partially observed or noise-corrupted data, using a partially observable Markov game (POMG) framework. We then propose a threefold heuristic solution procedure that: 1) uses the POMG to generate potential adversarial policies; 2) explicitly incorporates these adversarial policies in the construction of a robust defender policy by solving a robust dynamic program (DP); and 3) employs a probability matching heuristic in partially observable environments

Adversarial system · Adversary · Computer security · Heuristic · Machine learning · Markov chain · Markov decision process · Markov process · Mathematical optimization · Adversarial Robustness in Machine Learning · Computer Science · Infrastructure Resilience and Vulnerability Analysis · Mathematics · Reinforcement Learning in Robotics · Artificial Intelligence · Theoretical Computer Science

• Adaptation in Natural and Artificial Systems

  John H Holland•Adaptation in Natural and…•1992

  Add to list

• A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play

  Open Access•David Silver, Thomas Hubert et al.•Science•2018

  Add to list

• On Players′ Models of Other Players

  Open Access•Dale O Stahl, Paul W Wilson•Games and Economic Behavior•1995

  Add to list