Reproduce “Optimal Decentralized Smoothed Online Convex Optimization” 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
Neelkamal Bhuyan, Debankur Mukherjee, Adam Wierman
ACORD coordinates agents facing changing local costs, movement penalties, and costs for disagreement with neighbors. Agents repeatedly solve local optimization problems and exchange updates with immediate neighbors, without future predictions or sharing their hitting-cost functions. Under strongly convex hitting costs and squared-ℓ₂ switching costs, its competitive ratio approaches the lower bound ½ + ½√(1 + 4/μ_min) as iterations increase.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.