Keywords: CNN hardware acceleration, computation flow, compressed sparse filter format, 3D-SIMD processor architecture
TL;DR: We present a computation flow, stacked filters stationary flow, and a corresponding data encoding format, relative indexed compressed sparse filter format, for hardware acceleration of CNNs.
Abstract: To address memory and computation resource limitations for hardware-oriented acceleration of deep convolutional neural networks(CNNs), we present a computation flow, stacked filters stationary flow (SFS), and a corresponding data encoding format, relative indexed compressed sparse filter format (CSF), to make the best of data sparsity, and simplify data handling at execution time. Comparing with the state-of-the-art result (Han et al., 2016b), our methods achieve 1.11x improvement in reducing the storage required by AlexNet, and 1.09x improvement in reducing the storage required by SqueezeNet, without loss of accuracy on the ImageNet dataset. Moreover, using these approaches, chip area for logics handling irregular sparse data access can be saved. Comparing with the 2D-SIMD processure structures in DVAS, ENVISION, etc., our methods achieve about 3.65x processing element (PE) array utilization rate improvement (from 26.4% to 96.5%), using the data from Deep Compression on AlexNet.