Exploiting Low-Rank Tensor-Train Deep Neural Networks Based on Riemannian Gradient Descent With Illustrations of Speech Processing

TL;DR

This paper introduces low-rank tensor-train deep neural networks optimized with Riemannian gradient descent, combined with CNNs, to create efficient speech processing models that outperform traditional methods with fewer parameters.

Contribution

The work presents a novel hybrid tensor network model using Riemannian gradient descent for low-rank TT-DNNs, enhancing speech processing performance with reduced complexity.

Findings

LR-TT-DNN outperforms standard TT-DNN with fewer parameters.

CNN+(LR-TT-DNN) improves speech recognition accuracy.

Models achieve better results in speech enhancement and command recognition.

Abstract

This work focuses on designing low complexity hybrid tensor networks by considering trade-offs between the model complexity and practical performance. Firstly, we exploit a low-rank tensor-train deep neural network (TT-DNN) to build an end-to-end deep learning pipeline, namely LR-TT-DNN. Secondly, a hybrid model combining LR-TT-DNN with a convolutional neural network (CNN), which is denoted as CNN+(LR-TT-DNN), is set up to boost the performance. Instead of randomly assigning large TT-ranks for TT-DNN, we leverage Riemannian gradient descent to determine a TT-DNN associated with small TT-ranks. Furthermore, CNN+(LR-TT-DNN) consists of convolutional layers at the bottom for feature extraction and several TT layers at the top to solve regression and classification problems. We separately assess the LR-TT-DNN and CNN+(LR-TT-DNN) models on speech enhancement and spoken command recognition tasks. Our empirical evidence demonstrates that the LR-TT-DNN and CNN+(LR-TT-DNN) models with fewer model parameters can outperform the TT-DNN and CNN+(TT-DNN) counterparts.