Reproduce “Markov Chains Approximate Message Passing” 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
Amit Rajaraman, David X. Wu
The authors analyze recovery of a hidden ±1 signal from a noisy rank-one matrix using restricted Gaussian dynamics, whose correlation follows a scalar recursion related to approximate message passing. Under an SK mixing condition, they prove that for 1/λ<β<1, a uniformly selected Glauber iterate approaches a positive fixed point, but the guarantee requires e^{Ω(N^{4−2ε})} steps and is conditional, not a fast-time result.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.