PULP-NN: accelerating quantized neural networks on parallel ultra-low-power RISC-V processors

We present PULP-NN, an optimized computing library for a parallel ultra-low-power tightly coupled cluster of RISC-V processors. The key innovation in PULP-NN is a set of kernels for quantized neural network inference, targeting byte and subbyte data types, down to INT-1, tuned for the recent trend toward aggressive quantization in deep neural network inference. The proposed library exploits both the digital signal processing extensions available in the PULP RISC-V processors and the cluster’s parallelism, achieving up to 15.5 MACs/cycle on INT-8 and improving performance by up to 63× with respect to a sequential implementation on a single RISC-V core implementing the baseline RV32IMC ISA. Using PULP-NN, a CIFAR-10 network on an octa-core cluster runs in 30× and 19.6× less clock cycles than the current state-of-the-art ARM CMSISNN library, running on STM32L4 and STM32H7 MCUs, respectively. The proposed library, when running on a GAP-8 processor, outperforms by 36.8× and by 7.45× the execution on energy efficient MCUs such as STM32L4 and high-end MCUs such as STM32H7 respectively, when operating at the maximum frequency. The energy efficiency on GAP-8 is 14.1× higher than STM32L4 and 39.5× higher than STM32H7, at the maximum efficiency operating point. This article is part of the theme issue ‘Harmonizing energy-autonomous computing and intelligence’.