Researchers have developed a novel zeroth-order method to solve complex min-max problems involving non-smooth submodular-concave functions. This approach utilizes the subgradient of the Lovász extension for the minimizer and Gaussian smoothing for the maximizer to estimate function gradients. The algorithm demonstrates convergence to an epsilon-saddle point in the offline setting and achieves an O(sqrt(N*P_N)) online duality gap in the online setting, with theoretical results supported by numerical examples. AI
RANK_REASON This is a research paper detailing a new algorithmic approach to a specific class of optimization problems. [lever_c_demoted from research: ic=1 ai=0.4]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →