Workshop: New Frontiers in Adversarial Machine Learning

Reducing Exploitability with Population Based Training

Pavel Czempin ·


Self-play reinforcement learning has achieved state-of-the-art, and often superhuman, performance in a variety of zero-sum games.Yet prior work has found that policies that are highly capable against regular opponents can fail catastrophically against adversarial policies: an opponent trained explicitly against the victim.Prior defenses using adversarial training were able to make the victim robust to a specific adversary, but the victim remained vulnerable to new adversaries.We conjecture this limitation was due to insufficient diversity of adversaries seen during training.We propose a defense using population-based training to pit the victim against a range of opponents.We evaluate this defense's robustness against new adversaries in two low-dimensional environments.We find that our defense significantly increases robustness against adversaries in both environments and show that robustness is correlated with the size of the opponent population.

Chat is not available.