Tensor-Train Recurrent Neural Networks for Video Classification

TL;DR

This paper introduces a Tensor-Train RNN architecture that efficiently models high-dimensional sequential data like videos, achieving competitive results with simpler, more scalable models.

Contribution

The paper proposes a novel Tensor-Train factorization of RNN input-to-hidden weights, enabling effective high-dimensional sequence modeling without complex feature extractors.

Findings

Achieves competitive accuracy on video classification datasets.

Model complexity is significantly reduced compared to state-of-the-art.

Demonstrates the potential for transferring RNN architectures across domains.

Abstract

The Recurrent Neural Networks and their variants have shown promising performances in sequence modeling tasks such as Natural Language Processing. These models, however, turn out to be impractical and difficult to train when exposed to very high-dimensional inputs due to the large input-to-hidden weight matrix. This may have prevented RNNs' large-scale application in tasks that involve very high input dimensions such as video modeling; current approaches reduce the input dimensions using various feature extractors. To address this challenge, we propose a new, more general and efficient approach by factorizing the input-to-hidden weight matrix using Tensor-Train decomposition which is trained simultaneously with the weights themselves. We test our model on classification tasks using multiple real-world video datasets and achieve competitive performances with state-of-the-art models, even though our model architecture is orders of magnitude less complex. We believe that the proposed approach provides a novel and fundamental building block for modeling high-dimensional sequential data with RNN architectures and opens up many possibilities to transfer the expressive and advanced architectures from other domains such as NLP to modeling high-dimensional sequential data.