Matrix Completion with Model-free Weighting
This addresses the problem of handling highly varying missing patterns in matrix completion for applications like recommendation systems, though it appears incremental as it builds on existing work with improved theoretical bounds.
The paper tackles matrix completion under non-uniform missing data by proposing a model-free weighting method that adjusts for observation probabilities without explicit modeling, achieving stronger theoretical guarantees in scaling with respect to these probabilities under heterogeneous settings.
In this paper, we propose a novel method for matrix completion under general non-uniform missing structures. By controlling an upper bound of a novel balancing error, we construct weights that can actively adjust for the non-uniformity in the empirical risk without explicitly modeling the observation probabilities, and can be computed efficiently via convex optimization. The recovered matrix based on the proposed weighted empirical risk enjoys appealing theoretical guarantees. In particular, the proposed method achieves a stronger guarantee than existing work in terms of the scaling with respect to the observation probabilities, under asymptotically heterogeneous missing settings (where entry-wise observation probabilities can be of different orders). These settings can be regarded as a better theoretical model of missing patterns with highly varying probabilities. We also provide a new minimax lower bound under a class of heterogeneous settings. Numerical experiments are also provided to demonstrate the effectiveness of the proposed method.