Reproduce “CLS Hardness of 2D Min-Max Optimization with Product Constraints” 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
Alkis Kalavasis, Panagiotis Patsilinakos, Stratis Skoulakis, Manolis Zampetakis
The authors prove CLS-hardness for finding approximate projected gradient descent–ascent fixed points in a two-scalar min–max problem with independent constraints. They reduce End-of-Potential-Line to a smooth objective on the unit square. A separate deterministic black-box result gives an Ω(ε^-1/2) oracle-query lower bound for 100-smooth objectives. These results concern computational hardness and queries, not measured runtime or randomized algorithms.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.