1 Search Results for "Goel, Surbhi"

Document

DOI: 10.4230/LIPIcs.ITCS.2021.22

Tight Hardness Results for Training Depth-2 ReLU Networks

Authors: Surbhi Goel, Adam Klivans, Pasin Manurangsi, and Daniel Reichman

Published in: LIPIcs, Volume 185, 12th Innovations in Theoretical Computer Science Conference (ITCS 2021)

Abstract

We prove several hardness results for training depth-2 neural networks with the ReLU activation function; these networks are simply weighted sums (that may include negative coefficients) of ReLUs. Our goal is to output a depth-2 neural network that minimizes the square loss with respect to a given training set. We prove that this problem is NP-hard already for a network with a single ReLU. We also prove NP-hardness for outputting a weighted sum of k ReLUs minimizing the squared error (for k > 1) even in the realizable setting (i.e., when the labels are consistent with an unknown depth-2 ReLU network). We are also able to obtain lower bounds on the running time in terms of the desired additive error ε. To obtain our lower bounds, we use the Gap Exponential Time Hypothesis (Gap-ETH) as well as a new hypothesis regarding the hardness of approximating the well known Densest κ-Subgraph problem in subexponential time (these hypotheses are used separately in proving different lower bounds). For example, we prove that under reasonable hardness assumptions, any proper learning algorithm for finding the best fitting ReLU must run in time exponential in 1/ε². Together with a previous work regarding improperly learning a ReLU [Surbhi Goel et al., 2017], this implies the first separation between proper and improper algorithms for learning a ReLU. We also study the problem of properly learning a depth-2 network of ReLUs with bounded weights giving new (worst-case) upper bounds on the running time needed to learn such networks both in the realizable and agnostic settings. Our upper bounds on the running time essentially matches our lower bounds in terms of the dependency on ε.

Cite as

Surbhi Goel, Adam Klivans, Pasin Manurangsi, and Daniel Reichman. Tight Hardness Results for Training Depth-2 ReLU Networks. In 12th Innovations in Theoretical Computer Science Conference (ITCS 2021). Leibniz International Proceedings in Informatics (LIPIcs), Volume 185, pp. 22:1-22:14, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2021)

Copy BibTex To Clipboard

@InProceedings{goel_et_al:LIPIcs.ITCS.2021.22,
  author =	{Goel, Surbhi and Klivans, Adam and Manurangsi, Pasin and Reichman, Daniel},
  title =	{{Tight Hardness Results for Training Depth-2 ReLU Networks}},
  booktitle =	{12th Innovations in Theoretical Computer Science Conference (ITCS 2021)},
  pages =	{22:1--22:14},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-177-1},
  ISSN =	{1868-8969},
  year =	{2021},
  volume =	{185},
  editor =	{Lee, James R.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops-dev.dagstuhl.de/entities/document/10.4230/LIPIcs.ITCS.2021.22},
  URN =		{urn:nbn:de:0030-drops-135611},
  doi =		{10.4230/LIPIcs.ITCS.2021.22},
  annote =	{Keywords: ReLU, Learning Algorithm, Running Time Lower Bound}
}

Refine by Author
1 Goel, Surbhi
1 Klivans, Adam
1 Manurangsi, Pasin
1 Reichman, Daniel

Refine by Classification
1 Theory of computation → Machine learning theory
1 Theory of computation → Problems, reductions and completeness

Refine by Keyword
1 Learning Algorithm
1 ReLU
1 Running Time Lower Bound

Refine by Type
1 document

Refine by Publication Year
1 2021

1 Search Results for "Goel, Surbhi"

Tight Hardness Results for Training Depth-2 ReLU Networks

Abstract

Cite as

Thanks for your feedback!

Could not send message