Cambricon-X: An accelerator for sparse neural networks
- 1 October 2016
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
Neural networks (NNs) have been demonstrated to be useful in a broad range of applications such as image recognition, automatic translation and advertisement recommendation. State-of-the-art NNs are known to be both computationally and memory intensive, due to the ever-increasing deep structure, i.e., multiple layers with massive neurons and connections (i.e., synapses). Sparse neural networks have emerged as an effective solution to reduce the amount of computation and memory required. Though existing NN accelerators are able to efficiently process dense and regular networks, they cannot benefit from the reduction of synaptic weights. In this paper, we propose a novel accelerator, Cambricon-X, to exploit the sparsity and irregularity of NN models for increased efficiency. The proposed accelerator features a PE-based architecture consisting of multiple Processing Elements (PE). An Indexing Module (IM) efficiently selects and transfers needed neurons to connected PEs with reduced bandwidth requirement, while each PE stores irregular and compressed synapses for local computation in an asynchronous fashion. With 16 PEs, our accelerator is able to achieve at most 544 GOP/s in a small form factor (6.38 mm2 and 954 mW at 65 nm). Experimental results over a number of representative sparse networks show that our accelerator achieves, on average, 7.23x speedup and 6.43x energy saving against the state-of-the-art NN accelerator.Keywords
This publication has 26 references indexed in Scilit:
- ShiDianNaoPublished by Association for Computing Machinery (ACM) ,2015
- CaffePublished by Association for Computing Machinery (ACM) ,2014
- DianNaoACM SIGPLAN Notices, 2014
- Communication Optimization of Iterative Sparse Matrix-Vector Multiply on GPUs and FPGAsIEEE Transactions on Parallel and Distributed Systems, 2014
- A Massively Parallel Coprocessor for Convolutional Neural NetworksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Optimizing NUCA Organizations and Wiring Alternatives for Large Caches with CACTI 6.0Published by Institute of Electrical and Electronics Engineers (IEEE) ,2007
- The Impact of Arithmetic Representation on Implementing MLP-BP on FPGAs: A StudyIEEE Transactions on Neural Networks, 2007
- FPGA Implementation of a Pipelined On-Line BackpropagationJournal of Signal Processing Systems, 2005
- Recognition, Mining and SynthesisIntel Technology Journal, 2005
- Fat-trees: Universal networks for hardware-efficient supercomputingIEEE Transactions on Computers, 1985