Reproduce “Averaged Annealed Markov Chains Achieve the Algorithmic Threshold for Tensor PCA, Part I: Glauber Dynamics” 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
Aniket Das, Tselil Schramm
The authors introduce a two-phase local Glauber method for strong recovery in binary tensor PCA. It averages the states of a high-temperature chain to form a warm start, then uses fixed-temperature updates to amplify the signal. For every fixed k≥3, it guarantees recovery at λ≥Cβ,kN^(k−2)/4, with polynomial averaging; for even k, recovery is up to sign. This is not a mixing-time or posterior-sampling guarantee.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.