Genetic Adversarial Training of Decision Trees
We put forward a novel learning methodology for ensembles of decision trees based on a genetic algorithm that is able to train a decision tree for maximizing both its accuracy and its robustness to adversarial perturbations. This learning algorithm internally leverages a complete formal verification technique for robustness properties of decision trees based on abstract interpretation, a well-known static program analysis technique. We implemented this genetic adversarial training algorithm in a tool called MetaSilvae and we experimentally evaluated it on some standard reference datasets used in adversarial training. The experimental results show that MetaSilvae is able to train robust models that compete with and often improve on the current state-of-the-art of adversarial training of decision trees while being much more compact and therefore interpretable and efficient tree models.