Reproduce “Strategyproof Deep Reinforcement Learning for Online Resource Allocation” 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
Fariba Dehghan, Sebastian Stein, Vahid Yazdanpanah, Stephanie Gauthier
SP-SMM learns an online accept/reject rule for resource allocation that accommodates nonlinear interactions between task reports and occupied resources while preserving monotonicity. It prices accepted tasks at a verified, conservatively corrected acceptance threshold, yielding strategyproofness under the paper’s assumptions. On the four-resource Partner stream, it reaches 0.978 W/OPT versus 0.938 for LinearMono, against an offline LP upper bound.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.