Adversarial Machine Learning with Double Oracle

Adversarial Machine Learning with Double Oracle

Kai Wang

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
Doctoral Consortium. Pages 6472-6473. https://doi.org/10.24963/ijcai.2019/925

We aim to improve the general adversarial machine learning solution by introducing the double oracle idea from game theory, which is commonly used to solve a sequential zero-sum game, where the adversarial machine learning problem can be formulated as a zero-sum minimax problem between learner and attacker.
Keywords:
Machine Learning: Adversarial Machine Learning
Machine Learning: Ensemble Methods
Agent-based and Multi-agent Systems: Algorithmic Game Theory