Score: 2

Finding a dense submatrix of a random matrix. Sharp bounds for online algorithms

Published: July 25, 2025 | arXiv ID: 2507.19259v1

By: Shankar Bhamidi, David Gamarnik, Shuyang Gong

BigTech Affiliations: Massachusetts Institute of Technology

Potential Business Impact:

Finds hidden patterns in big data sets.

Business Areas:
A/B Testing Data and Analytics

We consider the problem of finding a dense submatrix of a matrix with i.i.d. Gaussian entries, where density is measured by average value. This problem arose from practical applications in biology and social sciences \cites{madeira-survey,shabalin2009finding} and is known to exhibit a computation-to-optimization gap between the optimal value and best values achievable by existing polynomial time algorithms. In this paper we consider the class of online algorithms, which includes the best known algorithm for this problem, and derive a tight approximation factor ${4\over 3\sqrt{2}}$ for this class. The result is established using a simple implementation of recently developed Branching-Overlap-Gap-Property \cite{huang2025tight}. We further extend our results to $(\mathbb R^n)^{\otimes p}$ tensors with i.i.d. Gaussian entries, for which the approximation factor is proven to be ${2\sqrt{p}/(1+p)}$.

Country of Origin
🇨🇳 🇺🇸 United States, China

Page Count
11 pages

Category
Mathematics:
Probability