English

Automatic Differentiation for Complex Valued SVD

Numerical Analysis 2019-11-07 v3 Statistical Mechanics Strongly Correlated Electrons Machine Learning Numerical Analysis Quantum Physics Machine Learning

Abstract

In this note, we report the back propagation formula for complex valued singular value decompositions (SVD). This formula is an important ingredient for a complete automatic differentiation(AD) infrastructure in terms of complex numbers, and it is also the key to understand and utilize AD in tensor networks.

Cite

@article{arxiv.1909.02659,
  title  = {Automatic Differentiation for Complex Valued SVD},
  author = {Zhou-Quan Wan and Shi-Xin Zhang},
  journal= {arXiv preprint arXiv:1909.02659},
  year   = {2019}
}

Comments

4.2 pages