Compressing Recurrent Neural Networks with Tensor Ring for Action Recognition
Open Access
- 17 July 2019
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 33 (01) , 4683-4690
Abstract
Recurrent Neural Networks (RNNs) and their variants, such as Long-Short Term Memory (LSTM) networks, and Gated Recurrent Unit (GRU) networks, have achieved promising performance in sequential data modeling. The hidden layers in RNNs can be regarded as the memory units, which are helpful in storing information in sequential contexts. However, when dealing with high dimensional input data, such as video and text, the input-to-hidden linear transformation in RNNs brings high memory usage and huge computational cost. This makes the training of RNNs very difficult. To address this challenge, we propose a novel compact LSTM model, named as TR-LSTM, by utilizing the low-rank tensor ring decomposition (TRD) to reformulate the input-to-hidden transformation. Compared with other tensor decomposition methods, TR-LSTM is more stable. In addition, TR-LSTM can complete an end-to-end training and also provide a fundamental building block for RNNs in handling large input data. Experiments on real-world action recognition datasets have demonstrated the promising performance of the proposed TR-LSTM compared with the tensor-train LSTM and other state-of-the-art competitors.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- TensorD: A tensor decomposition library in TensorFlowNeurocomputing, 2018
- NeuralCP: Bayesian Multiway Data Analysis with Neural Tensor DecompositionCognitive Computation, 2018
- Knowledge Base Completion by Variational Bayesian Neural Tensor DecompositionCognitive Computation, 2018
- Scene labeling with LSTM recurrent neural networks2015
- Multi-view Super Vector for Action Recognition2014
- Incremental Activity Modeling and Recognition in Streaming Videos2014
- Robust action recognition using local motion and group sparsityPattern Recognition, 2014
- Better Exploiting Motion for Better Action Recognition2013
- Action recognition by dense trajectories2011
- Object, Scene and Actions: Combining Multiple Features for Human Action Recognition2010