Benchmark Comparisons of Spike-based Reconfigurable Neuroprocessor Architectures for Control Applications

• DOI: 10.1145/3526241.3530381
• Publication Date: 2022-06-02T14:37:09Z
• Authors: Adam Z. Foshie; Charles Rizzo; Hritom Das; Chaohui Zheng; James S. Plank; Garrett S. Rose
• Keywords: Benchmark (surveying); Neuromorphic engineering

Neuromorphic computing is a leading option for non von-Neumann architectures. With it, neural networks are developed that derive architectural inspiration from how the brain operates with neurons, synapses, and spikes. These often implemented in either software or hardware based neuroprocessors designed to handle specific tasks efficiently. Even if hardware, emulation instrumental determining worthwhile features capabilities of architecture. In this work two novel introduced: software-based RISP neuroprocessor, RAVENS neuroprocessor. Several benchmark tests using control applications performed each neuroprocessor configured various ways evaluate their comparative performance training properties.