Details
Poster
Presenter(s)
Display Name
Zihan Xia
- Affiliation
-
AffiliationUniversity of Electronic Science and Technology of China
- Country
Abstract
We proposed the neural synaptic plasticity-like computing (NSPC) to simulate the neural network activity for inference task with ultra-simple logic gates. To this end, the NSPC imitates the structure of the neural synaptic plasticity from a circuit wires connection perspective. The multiplication of weight in traditional ANNs is transformed by the wire connectivity in NSPC, which requires only bundle of wires without any logics. The proposed NSPC exhibits comparable accuracy with low cost performance. According to implementation results, the NSPC requires only 28% hardware cost of conventional ANNs scheme and with 114% throughput improvement on the average.