TY - JOUR AU - Einziger, Gil AU - Goldstein, Maayan AU - Sa’ar, Yaniv AU - Segall, Itai PY - 2019/07/17 Y2 - 2024/03/28 TI - Verifying Robustness of Gradient Boosted Models JF - Proceedings of the AAAI Conference on Artificial Intelligence JA - AAAI VL - 33 IS - 01 SE - AAAI Technical Track: Human-AI Collaboration DO - 10.1609/aaai.v33i01.33012446 UR - https://ojs.aaai.org/index.php/AAAI/article/view/4089 SP - 2446-2453 AB - <p>Gradient boosted models are a fundamental machine learning technique. Robustness to small perturbations of the input is an important quality measure for machine learning models, but the literature lacks a method to prove the robustness of gradient boosted models.</p><p>This work introduces V<sc>ERI</sc>GB, a tool for quantifying the robustness of gradient boosted models. V<sc>ERI</sc>GB encodes the model and the robustness property as an SMT formula, which enables state of the art verification tools to prove the model’s robustness. We extensively evaluate V<sc>ERI</sc>GB on publicly available datasets and demonstrate a capability for verifying large models. Finally, we show that some model configurations tend to be inherently more robust than others.</p> ER -