Shenjing: A low power reconfigurable neuromorphic accelerator with partial-sum and spike networks-on-chip
This work addresses the need for low-power, on-device AI accelerators by enabling efficient mapping of conventional neural networks to SNNs without model modification, which is incremental but practical for edge computing applications.
The authors tackled the problem of energy-efficient on-device AI by proposing Shenjing, a reconfigurable neuromorphic accelerator for spiking neural networks (SNNs) that eliminates multipliers and exposes on-chip communications to software, achieving 96% accuracy on MNIST with a multilayer perceptron while consuming only 1.26mW using 10 cores.
The next wave of on-device AI will likely require energy-efficient deep neural networks. Brain-inspired spiking neural networks (SNN) has been identified to be a promising candidate. Doing away with the need for multipliers significantly reduces energy. For on-device applications, besides computation, communication also incurs a significant amount of energy and time. In this paper, we propose Shenjing, a configurable SNN architecture which fully exposes all on-chip communications to software, enabling software mapping of SNN models with high accuracy at low power. Unlike prior SNN architectures like TrueNorth, Shenjing does not require any model modification and retraining for the mapping. We show that conventional artificial neural networks (ANN) such as multilayer perceptron, convolutional neural networks, as well as the latest residual neural networks can be mapped successfully onto Shenjing, realizing ANNs with SNN's energy efficiency. For the MNIST inference problem using a multilayer perceptron, we were able to achieve an accuracy of 96% while consuming just 1.26mW using 10 Shenjing cores.