Interaction Decompositions for Tensor Network Regression

Abstract

It is well known that tensor network regression models operate on an exponentially large feature space, but questions remain as to how effectively they are able to utilize this space. Using a polynomial featurization, we propose the interaction decomposition as a tool that can assess the relative importance of different regressors as a function of their polynomial degree. We apply this decomposition to tensor ring and tree tensor network models trained on the MNIST and Fashion MNIST datasets, and find that up to 75% of interaction degrees are contributing meaningfully to these models. We also introduce a new type of tensor network model that is explicitly trained on only a small subset of interaction degrees, and find that these models are able to match or even outperform the full models using only a fraction of the exponential feature space. This suggests that standard tensor network models utilize their polynomial regressors in an inefficient manner, with the lower degree terms being vastly under-utilized.

arXiv comment22 pages, 3 figures

References in corpus (11)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• The density-matrix renormalization group in the age of matrix product states
• Classical simulation of quantum many-body systems with a tree tensor network
• Hand-waving and Interpretive Dance: An Introductory Course on Tensor Networks
• Tensor Ring Decomposition
• Tensor network states and geometry
• Simulating Strongly Correlated Quantum Systems with Tree Tensor Networks
• Tensor Networks in a Nutshell
• Machine Learning by Unitary Tensor Network of Hierarchical Tree Structure
• Learning Relevant Features of Data with Multi-scale Tensor Networks
• TensorNetwork for Machine Learning

Cited by in corpus (1)

• Machine learning with tree tensor networks, CP rank constraints, and tensor dropout