English

Learning Unitary Operators with Help From u(n)

Machine Learning 2017-01-11 v3 Machine Learning

Abstract

A major challenge in the training of recurrent neural networks is the so-called vanishing or exploding gradient problem. The use of a norm-preserving transition operator can address this issue, but parametrization is challenging. In this work we focus on unitary operators and describe a parametrization using the Lie algebra u(n)\mathfrak{u}(n) associated with the Lie group U(n)U(n) of n×nn \times n unitary matrices. The exponential map provides a correspondence between these spaces, and allows us to define a unitary matrix using n2n^2 real coefficients relative to a basis of the Lie algebra. The parametrization is closed under additive updates of these coefficients, and thus provides a simple space in which to do gradient descent. We demonstrate the effectiveness of this parametrization on the problem of learning arbitrary unitary operators, comparing to several baselines and outperforming a recently-proposed lower-dimensional parametrization. We additionally use our parametrization to generalize a recently-proposed unitary recurrent neural network to arbitrary unitary matrices, using it to solve standard long-memory tasks.

Keywords

Cite

@article{arxiv.1607.04903,
  title  = {Learning Unitary Operators with Help From u(n)},
  author = {Stephanie L. Hyland and Gunnar Rätsch},
  journal= {arXiv preprint arXiv:1607.04903},
  year   = {2017}
}

Comments

9 pages, 3 figures, 5 figures inc. subfigures, to appear at AAAI-17

R2 v1 2026-06-22T14:56:46.741Z