The Fifth International Verification of Neural Networks Competition (VNN-COMP 2024): Summary and Results

TL;DR

The paper reports on the 5th VNN-COMP, an annual competition that benchmarks neural network verification tools, highlighting standardization efforts, diverse benchmarks, and community engagement.

Contribution

It introduces standardized formats and evaluation procedures for neural network verification tools, fostering fair comparison and progress in the field.

Findings

8 teams participated with diverse tools

Evaluation on 20 benchmarks using standardized formats

Insights gained for future tool development

Abstract

This report summarizes the 5th International Verification of Neural Networks Competition (VNN-COMP 2024), held as a part of the 7th International Symposium on AI Verification (SAIV), that was collocated with the 36th International Conference on Computer-Aided Verification (CAV). VNN-COMP is held annually to facilitate the fair and objective comparison of state-of-the-art neural network verification tools, encourage the standardization of tool interfaces, and bring together the neural network verification community. To this end, standardized formats for networks (ONNX) and specification (VNN-LIB) were defined, tools were evaluated on equal-cost hardware (using an automatic evaluation pipeline based on AWS instances), and tool parameters were chosen by the participants before the final test sets were made public. In the 2024 iteration, 8 teams participated on a diverse set of 12 regular and 8 extended benchmarks. This report summarizes the rules, benchmarks, participating tools, results, and lessons learned from this iteration of this competition.