ML-Dev-Bench: Comparative Analysis of AI Agents on ML development workflows
This work addresses the need for better benchmarks to assess AI agents in practical ML development for researchers and practitioners, though it is incremental as it builds on existing benchmarks by expanding scope.
The authors tackled the problem of evaluating AI agents on full machine learning development workflows by introducing ML-Dev-Bench, a benchmark that tests capabilities across 30 tasks including dataset handling and debugging, and found insights into the strengths and limitations of three agents (ReAct, Openhands, AIDE).
In this report, we present ML-Dev-Bench, a benchmark aimed at testing agentic capabilities on applied Machine Learning development tasks. While existing benchmarks focus on isolated coding tasks or Kaggle-style competitions, ML-Dev-Bench tests agents' ability to handle the full complexity of ML development workflows. The benchmark assesses performance across critical aspects including dataset handling, model training, improving existing models, debugging, and API integration with popular ML tools. We evaluate three agents - ReAct, Openhands, and AIDE - on a diverse set of 30 tasks, providing insights into their strengths and limitations in handling practical ML development challenges. We open source the benchmark for the benefit of the community at \href{https://github.com/ml-dev-bench/ml-dev-bench}{https://github.com/ml-dev-bench/ml-dev-bench}.