Compressing Recurrent Neural Network with Tensor Train
This work addresses the computational resource demands of RNNs for researchers and practitioners, though it is incremental as it applies an existing compression method to RNN architectures.
The paper tackles the problem of large parameter sizes in Recurrent Neural Networks (RNNs) by proposing a model that compresses weights using the Tensor Train format, achieving up to 40 times parameter reduction while preserving performance on sequence tasks.
Recurrent Neural Network (RNN) are a popular choice for modeling temporal and sequential tasks and achieve many state-of-the-art performance on various complex problems. However, most of the state-of-the-art RNNs have millions of parameters and require many computational resources for training and predicting new data. This paper proposes an alternative RNN model to reduce the number of parameters significantly by representing the weight parameters based on Tensor Train (TT) format. In this paper, we implement the TT-format representation for several RNN architectures such as simple RNN and Gated Recurrent Unit (GRU). We compare and evaluate our proposed RNN model with uncompressed RNN model on sequence classification and sequence prediction tasks. Our proposed RNNs with TT-format are able to preserve the performance while reducing the number of RNN parameters significantly up to 40 times smaller.