Neuromorphic Algorithm-hardware Codesign for Temporal Pattern Learning
Hosted in Virtual Platform
Emerging Models of Computation
DescriptionNeuromorphic computing and spiking neural network (SNN) have drawn interests as the potential to perform cognitive tasks with high energy efficiency. However, some factors such as temporal dynamics and spike timings are often ignored by existing works, limiting the performance and application of neuromorphic computing. Firstly, due to the lack of effective SNN training algorithm, it is difficult to utilize the temporal neural dynamics. Many existing algorithms still treat neuron activation statistically. Secondly, utilizing temporal neural dynamics poses challenges on hardware design, In this work, we proposed a neuromorphic hardware algorithm-hardware codesign to learn temporal patterns.