Sketched Ridge Regression: Optimization Perspective, Statistical Perspective, and Model Averaging

Abstract

We address the statistical and optimization impacts of the classical sketch and Hessian sketch used to approximately solve the Matrix Ridge Regression (MRR) problem. Prior research has quantified the effects of classical sketch on the strictly simpler least squares regression (LSR) problem. We establish that classical sketch has a similar effect upon the optimization properties of MRR as it does on those of LSR: namely, it recovers nearly optimal solutions. By contrast, Hessian sketch does not have this guarantee, instead, the approximation error is governed by a subtle interplay between the "mass" in the responses and the optimal objective value. For both types of approximation, the regularization in the sketched MRR problem results in significantly different statistical properties from those of the sketched LSR problem. In particular, there is a bias-variance trade-off in sketched MRR that is not present in sketched LSR. We provide upper and lower bounds on the bias and variance of sketched MRR, these bounds show that classical sketch significantly increases the variance, while Hessian sketch significantly increases the bias. Empirically, sketched MRR solutions can have risks that are higher by an order-of-magnitude than those of the optimal MRR solutions. We establish theoretically and empirically that model averaging greatly decreases the gap between the risks of the true and sketched solutions to the MRR problem. Thus, in parallel or distributed settings, sketching combined with model averaging is a powerful technique that quickly obtains near-optimal solutions to the MRR problem while greatly mitigating the increased statistical risk incurred by sketching.

To appear in Journal of Machine Learning Research, 2018. A short version has appeared in International Conference on Machine Learning (ICML), 2017

References in corpus (4)

• Iterative Hessian sketch: Fast and accurate solution approximation for constrained least-squares
• Unbiased estimates for linear regression via volume sampling
• OSNAP: Faster numerical linear algebra algorithms via sparser subspace embeddings
• Subsampling for Ridge Regression via Regularized Volume Sampling

Cited by in corpus (16)

• How to reduce dimension with PCA and random projections?
• Randomized spectral co-clustering for large-scale directed networks
• Distributed Sketching Methods for Privacy Preserving Regression
• Sketchy Empirical Natural Gradient Methods for Deep Learning
• Data-driven Random Fourier Features using Stein Effect
• Distributed Averaging Methods for Randomized Second Order Optimization
• A Deterministic Streaming Sketch for Ridge Regression
• Large-Scale Beamforming for Massive MIMO via Randomized Sketching
• Sketching in Bayesian High Dimensional Regression With Big Data Using Gaussian Scale Mixture Priors
• Effective Dimension Adaptive Sketching Methods for Faster Regularized Least-Squares Optimization
• Ridge Regression with Frequent Directions: Statistical and Optimization Perspectives
• Fast Convex Quadratic Optimization Solvers with Adaptive Sketching-based Preconditioners
• Learning Augmentation Distributions using Transformed Risk Minimization
• Sparse sketches with small inversion bias
• $\text{RidgeSketch}$: A Fast sketching based solver for large scale ridge regression
• NoisyCUR: An algorithm for two-cost budgeted matrix completion