Reproduce “Finite-Time Minimax Bounds and an Optimal Lyapunov Policy in Queueing Control” 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
Yujie Liu, Vincent Y. F. Tan, Yunbei Xu
LyapOpt schedules parallel queues by minimizing the squared backlog predicted to remain after service, rather than maximizing backlog-weighted service. The paper studies expected total queue length at a horizon and establishes a minimax benchmark. Under a per-slot dominated-region condition, its bound matches the leading heavy-traffic lower-bound term up to constants; a deterministic two-queue example gives a finite-range √T lower bound for MaxWeight.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.