PIRSA:18050061

Video URL

https://pirsa.org/18050061

Classical and Quantum Machine Learning with Tensor Networks

Stoudenmire, M. (2018). Classical and Quantum Machine Learning with Tensor Networks. Perimeter Institute for Theoretical Physics. https://pirsa.org/18050061

Stoudenmire, Miles. Classical and Quantum Machine Learning with Tensor Networks. Perimeter Institute for Theoretical Physics, May. 18, 2018, https://pirsa.org/18050061 

          @misc{ scivideos_PIRSA:18050061,
            doi = {10.48660/18050061},
            url = {https://pirsa.org/18050061},
            author = {Stoudenmire, Miles},
            keywords = {Quantum Matter},
            language = {en},
            title = {Classical and Quantum Machine Learning with Tensor Networks},
            publisher = {Perimeter Institute for Theoretical Physics},
            year = {2018},
            month = {may},
            note = {PIRSA:18050061 see, \url{https://scivideos.org/index.php/pirsa/18050061}}
          }

Miles Stoudenmire Flatiron Institute

Talk numberPIRSA:18050061
DOI10.48660/18050061
Source RepositoryPIRSA
Collection
Talk Type Scientific Series
Subject

Abstract

Over the last decade, there have been enormous gains in machine learning technology primarily driven by neural networks. A major reason neural networks have outperformed older techniques is that the cost of optimizing them scales well with the size of the training dataset. But neural networks have the drawback that they are not very well understood theoretically.

 

Recent work by several groups has explored an alternative approach to creating machine learning model functions based on tensor networks, which are a technique for parameterizing many-body quantum wavefunctions. The cost of training tensor network models scales similarly to the cost of training neural networks. In addition, their relatively simple, linear structure has provided good theoretical understanding of their properties, and underpins many powerful techniques to optimize and manipulate them. 

 

After introducing tensor network machine learning models, I will discuss some of the techniques to optimize them and results for supervised and generative machine learning tasks. These techniques can automatically adapt the number of parameters and suggest interesting interpretations and extensions for 'deep' tensor network variants. I will conclude by discussing a recent tensor network based proposal to formulate hybrid quantum-classical algorithms for machine learning with quantum computers.