English

Provable Bounds for Learning Some Deep Representations

Machine Learning 2013-10-25 v1 Artificial Intelligence Machine Learning

Abstract

We give algorithms with provable guarantees that learn a class of deep nets in the generative model view popularized by Hinton and others. Our generative model is an nn node multilayer neural net that has degree at most nγn^{\gamma} for some γ<1\gamma <1 and each edge has a random edge weight in [1,1][-1,1]. Our algorithm learns {\em almost all} networks in this class with polynomial running time. The sample complexity is quadratic or cubic depending upon the details of the model. The algorithm uses layerwise learning. It is based upon a novel idea of observing correlations among features and using these to infer the underlying edge structure via a global graph recovery procedure. The analysis of the algorithm reveals interesting structure of neural networks with random edge weights.

Keywords

Cite

@article{arxiv.1310.6343,
  title  = {Provable Bounds for Learning Some Deep Representations},
  author = {Sanjeev Arora and Aditya Bhaskara and Rong Ge and Tengyu Ma},
  journal= {arXiv preprint arXiv:1310.6343},
  year   = {2013}
}

Comments

The first 18 pages serve as an extended abstract and a 36 pages long technical appendix follows

R2 v1 2026-06-22T01:52:45.574Z