Exploiting and Enhancing Computation Latency Variability for High-Performance Time-Domain Computing-in-Memory Neural Network Accelerators.
Chia-Chun WangYun-Chen LoJun-Shen WuYu-Chih TsaiChia-Cheng ChangTsen-Wei HsuMin-Wei ChuChuan-Yao LaiRen-Shuo LiuPublished in: ICCD (2023)
Keyphrases
- neural network
- efficiently computing
- auto associative
- efficient computation
- low latency
- data transfer
- back propagation
- frequency domain
- parallel computers
- genetic algorithm
- computing power
- memory bandwidth
- low overhead
- memory requirements
- recurrent neural networks
- main memory
- fault diagnosis
- fuzzy logic
- artificial neural networks
- neural network model
- parallel computation
- response time
- general purpose