Query Efficient Structured Matrix Learning

We study the problem of learning a structured approximation (low-rank, sparse, banded, etc.) to an unknown matrix A given access to matrix-vector product (matvec) queries of the form x rightarrow Ax and x rightarrow ATx. This problem is of central importance to algorithms across scientific computing and machine learning, with applications to fast multiplication and inversion for structured matrices, building preconditioners for first-order optimization, and as a model for differential operator learning. Prior work focuses on obtaining query complexity upper and lower bounds for learning specific structured matrix families that commonly arise in applications. We initiate the study of the problem in greater generality, aiming to understand the query complexity of learning approximations from general matrix families. Our main result focuses on finding a near-optimal approximation to A from any finite-sized family of matrices, F. Standard results from matrix sketching show that O(log|F|) matvec queries suffice in this setting. This bound can also be achieved, and is optimal, for vector-matrix-vector queries of the form x,yrightarrow xTAy, which have been widely studied in work on rank-1 matrix sensing. Surprisingly, we show that, in the matvec model, it is possible to obtain a nearly quadratic improvement in complexity, to tildeO(√log|F|). Further, we prove that this bound is tight up to log-log factors. Via covering number arguments, our result extends to well-studied infinite families. As an example, we establish that a near-optimal approximation from any linear matrix family of dimension q can be learned with tildeO(√q) matvec queries, improving on an O(q) bound achievable via sketching techniques and vector-matrix-vector queries.