License: Creative Commons Attribution 3.0 Unported license (CC BY 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.APPROX/RANDOM.2020.50
URN: urn:nbn:de:0030-drops-126532
Go to the corresponding LIPIcs Volume Portal

Andoni, Alexandr ; Burns, Collin ; Li, Yi ; Mahabadi, Sepideh ; Woodruff, David P.

Streaming Complexity of SVMs

LIPIcs-APPROX50.pdf (0.6 MB)


We study the space complexity of solving the bias-regularized SVM problem in the streaming model. In particular, given a data set (x_i,y_i) ∈ ℝ^d× {-1,+1}, the objective function is F_λ(θ,b) = λ/2‖(θ,b)‖₂² + 1/n∑_{i=1}ⁿ max{0,1-y_i(θ^Tx_i+b)} and the goal is to find the parameters that (approximately) minimize this objective. This is a classic supervised learning problem that has drawn lots of attention, including for developing fast algorithms for solving the problem approximately: i.e., for finding (θ,b) such that F_λ(θ,b) ≤ min_{(θ',b')} F_λ(θ',b')+ε.
One of the most widely used algorithms for approximately optimizing the SVM objective is Stochastic Gradient Descent (SGD), which requires only O(1/λε) random samples, and which immediately yields a streaming algorithm that uses O(d/λε) 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_λ on any query (θ, b). We show that, for both problems, for dimensions d = 1,2, one can obtain streaming algorithms with space polynomially smaller than 1/λε, which is the complexity of SGD for strongly convex functions like the bias-regularized SVM [Shalev-Shwartz et al., 2007], and which is known to be tight in general, even for d = 1 [Agarwal et al., 2009]. We also prove polynomial lower bounds for both point estimation and optimization. In particular, for point estimation we obtain a tight bound of Θ(1/√{ε}) for d = 1 and a nearly tight lower bound of Ω̃(d/{ε}²) for d = Ω(log(1/ε)). Finally, for optimization, we prove a Ω(1/√{ε}) lower bound for d = Ω(log(1/ε)), and show similar bounds when d is constant.

BibTeX - Entry

  author =	{Alexandr Andoni and Collin Burns and Yi Li and Sepideh Mahabadi and David P. Woodruff},
  title =	{{Streaming Complexity of SVMs}},
  booktitle =	{Approximation, Randomization, and Combinatorial Optimization. Algorithms and Techniques (APPROX/RANDOM 2020)},
  pages =	{50:1--50:22},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-164-1},
  ISSN =	{1868-8969},
  year =	{2020},
  volume =	{176},
  editor =	{Jaros{\l}aw Byrka and Raghu Meka},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{},
  URN =		{urn:nbn:de:0030-drops-126532},
  doi =		{10.4230/LIPIcs.APPROX/RANDOM.2020.50},
  annote =	{Keywords: support vector machine, streaming algorithm, space lower bound, sketching algorithm, point estimation}

Keywords: support vector machine, streaming algorithm, space lower bound, sketching algorithm, point estimation
Collection: Approximation, Randomization, and Combinatorial Optimization. Algorithms and Techniques (APPROX/RANDOM 2020)
Issue Date: 2020
Date of publication: 11.08.2020

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI