Benchmarks in GenBaB

This repository contains benchmarks used in GenBaB proposed in the following paper:

Zhouxing Shi*, Qirui Jin*, Zico Kolter, Suman Jana, Cho-Jui Hsieh, Huan Zhang. Neural Network Verification with Branch-and-Bound for General Nonlinearities. To appear in TACAS 2025. (*Equal contribution)

Acknowledgement

• ml4acopf was generated from AI4OPT/ml4acopf_benchmark.
• eran was from ERAN.