English

Sample Complexity Bounds for Recurrent Neural Networks with Application to Combinatorial Graph Problems

Machine Learning 2019-11-20 v2 Machine Learning

Abstract

Learning to predict solutions to real-valued combinatorial graph problems promises efficient approximations. As demonstrated based on the NP-hard edge clique cover number, recurrent neural networks (RNNs) are particularly suited for this task and can even outperform state-of-the-art heuristics. However, the theoretical framework for estimating real-valued RNNs is understood only poorly. As our primary contribution, this is the first work that upper bounds the sample complexity for learning real-valued RNNs. While such derivations have been made earlier for feed-forward and convolutional neural networks, our work presents the first such attempt for recurrent neural networks. Given a single-layer RNN with aa rectified linear units and input of length bb, we show that a population prediction error of ε\varepsilon can be realized with at most O~(a4b/ε2)\tilde{\mathcal{O}}(a^4b/\varepsilon^2) samples. We further derive comparable results for multi-layer RNNs. Accordingly, a size-adaptive RNN fed with graphs of at most nn vertices can be learned in O~(n6/ε2)\tilde{\mathcal{O}}(n^6/\varepsilon^2), i.e., with only a polynomial number of samples. For combinatorial graph problems, this provides a theoretical foundation that renders RNNs competitive.

Keywords

Cite

@article{arxiv.1901.10289,
  title  = {Sample Complexity Bounds for Recurrent Neural Networks with Application to Combinatorial Graph Problems},
  author = {Nil-Jana Akpinar and Bernhard Kratzwald and Stefan Feuerriegel},
  journal= {arXiv preprint arXiv:1901.10289},
  year   = {2019}
}

Comments

A two-page summary of this paper has been accepted as a student abstract at AAAI-20, this is the extended full version

R2 v1 2026-06-23T07:25:35.108Z