Versatile Verification of Tree Ensembles

TL;DR

Veritas is a versatile algorithm for verifying various properties of tree ensemble models, offering efficient bounds and solutions that outperform previous methods in speed and accuracy, enabling more complex real-world verification tasks.

Contribution

Introduces Veritas, a generic, efficient verification algorithm for tree ensembles that handles multiple tasks with bounds and solutions, surpassing prior specialized approaches.

Findings

Outperforms previous methods in solution accuracy and speed

Provides tighter bounds and more frequent exact solutions

Enables verification of larger, real-world models

Abstract

Machine learned models often must abide by certain requirements (e.g., fairness or legal). This has spurred interested in developing approaches that can provably verify whether a model satisfies certain properties. This paper introduces a generic algorithm called Veritas that enables tackling multiple different verification tasks for tree ensemble models like random forests (RFs) and gradient boosting decision trees (GBDTs). This generality contrasts with previous work, which has focused exclusively on either adversarial example generation or robustness checking. Veritas formulates the verification task as a generic optimization problem and introduces a novel search space representation. Veritas offers two key advantages. First, it provides anytime lower and upper bounds when the optimization problem cannot be solved exactly. In contrast, many existing methods have focused on exact solutions and are thus limited by the verification problem being NP-complete. Second, Veritas produces full (bounded suboptimal) solutions that can be used to generate concrete examples. We experimentally show that Veritas outperforms the previous state of the art by (a) generating exact solutions more frequently, (b) producing tighter bounds when (a) is not possible, and (c) offering orders of magnitude speed ups. Subsequently, Veritas enables tackling more and larger real-world verification scenarios.