Streaming Complexity of SVMs

TL;DR

This paper investigates the space complexity of streaming algorithms for bias-regularized SVMs, revealing polynomial lower bounds and more space-efficient algorithms for low-dimensional cases, advancing understanding of streaming SVM optimization.

Contribution

It introduces new space complexity bounds for streaming SVM algorithms, including polynomial lower bounds and algorithms that outperform SGD in low dimensions.

Findings

Polynomial space lower bounds for point estimation and optimization.

Streaming algorithms with space polynomially smaller than SGD for low dimensions.

Tight bounds of (1/\u03b5) for point estimation in 1D.

Abstract

We study the space complexity of solving the bias-regularized SVM problem in the streaming model. This is a classic supervised learning problem that has drawn lots of attention, including for developing fast algorithms for solving the problem approximately. One of the most widely used algorithms for approximately optimizing the SVM objective is Stochastic Gradient Descent (SGD), which requires only $O(\frac{1}{\lambda\epsilon})$ random samples, and which immediately yields a streaming algorithm that uses $O(\frac{d}{\lambda\epsilon})$ space. For related problems, better streaming algorithms are only known for smooth functions, unlike the SVM objective that we focus on in this work. We initiate an investigation of the space complexity for both finding an approximate optimum of this objective, and for the related ``point estimation'' problem of sketching the data set to evaluate the function value $F_\lambda$ on any query $(\theta, b)$. We show that, for both problems, for dimensions $d=1,2$, one can obtain streaming algorithms with space polynomially smaller than $\frac{1}{\lambda\epsilon}$, which is the complexity of SGD for strongly convex functions like the bias-regularized SVM, and which is known to be tight in general, even for $d=1$. We also prove polynomial lower bounds for both point estimation and optimization. In particular, for point estimation we obtain a tight bound of $\Theta(1/\sqrt{\epsilon})$ for $d=1$ and a nearly tight lower bound of $\widetilde{\Omega}(d/{\epsilon}^2)$ for $d = \Omega( \log(1/\epsilon))$. Finally, for optimization, we prove a $\Omega(1/\sqrt{\epsilon})$ lower bound for $d = \Omega( \log(1/\epsilon))$, and show similar bounds when $d$ is constant.