1. Introduction
This paper is published in ISSCC in 2017. Recently, deep learning with convolutional neural networks (CNNs) and recurrent neural networks (RNNs) has become universal in all-around applications. But there has been no work on a combined CNN-RNN processor. Hence, the researchs present a reconfigurable CNN-RNN processor.
2. Innovation point
The computational requirements in CNNs are quite different from those of RNNs, as shown in the following figure.
While convolution layers require a massive amount of computation with a relatively small number of filter weights, fully-connected layers and RNN-LSTM layers require a relatively small amount of computation with a huge number of filter weights. Thus, a combined CNN-RNN processor with high energy-efficiency is implemented. The following is the overall architecture of DNPU.
3. Summary
The chip photograph and specifications are shown below.