Reproduce “When are Overcomplete Topic Models Identifiable? Uniqueness of Tensor Tucker Decompositions with Structured Sparsity” 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
Animashree Anandkumar, Daniel Hsu, Majid Janzamin, Sham Kakade
The authors analyze when an overcomplete topic model can be identified if one hidden topic persists across blocks of words. This persistence gives higher-order observed moments a structure that can distinguish topics, provided support-matching, rank, and non-degeneracy conditions hold. For n≥2, generic recovery is possible from exact moments of order at least 2n; a random-support result reaches q=O(p^n), under stated conditions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.