Reproduce “OptScientist: Automated Discovery of Optimization Laws” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Ma ZhenYao, Yue Liang, Jian-Qiao Zhu

OptScientist learns optimization representations from observed contexts and outcomes, then distills objectives and constraints into compact symbolic laws. On OptFeynman, it perfectly recovers preference rankings on 88 of 99 discrete and 78 of 99 continuous tasks. For central-bank policy, it reaches 78.6% exact stance accuracy and 85.7% direction accuracy on 14 future quarters, supporting prediction, not proof of policymakers’ decision process.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.