Learning-Based Low-Rank Approximations

Abstract

We introduce a "learning-based" algorithm for the low-rank decomposition problem: given an $n\times d$ matrix $A$, and a parameter $k$, compute a rank-$k$ matrix $A^{\prime }$ that minimizes the approximation loss $\Vert A-A^{\prime }{\Vert }_F$. The algorithm uses a training set of input matrices in order to optimize its performance. Specifically, some of the most efficient approximate algorithms for computing low-rank approximations proceed by computing a projection $SA$, where $S$ is a sparse random $m\times n$ "sketching matrix", and then performing the singular value decomposition of $SA$. We show how to replace the random matrix $S$ with a "learned" matrix of the same sparsity to reduce the error. Our experiments show that, for multiple types of data sets, a learned sketch matrix can substantially reduce the approximation loss compared to a random matrix $S$, sometimes by one order of magnitude. We also study mixed matrices where only some of the rows are trained and the remaining ones are random, and show that matrices still offer improved performance while retaining worst-case guarantees. Finally, to understand the theoretical aspects of our approach, we study the special case of $m=1$. In particular, we give an approximation algorithm for minimizing the empirical loss, with approximation factor depending on the stable rank of matrices in the training set. We also show generalization bounds for the sketch matrix learning problem.

NeurIPS 2019

Cited by in corpus (9)

• Learning to Optimize: A Primer and A Benchmark
• CPGD: Cadzow Plug-and-Play Gradient Descent for Generalised FRI
• Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms
• (Learned) Frequency Estimation Algorithms under Zipfian Distribution
• Sparse Linear Networks with a Fixed Butterfly Structure: Theory and Practice
• Composable Sketches for Functions of Frequencies: Beyond the Worst Case
• Learning-Augmented Sketches for Hessians
• Learning the Positions in CountSketch
• Learning Based Proximity Matrix Factorization for Node Embedding