Charge Based Mixed-Signal Multiply-Accumulate Circuit for Energy Efficient In-Memory Computing

Raphael Nägele,Felix Wiewel, Sebastian Kelz,Manuel Wittlinger,Manfred Berroth,Bin Yang,Markus Grözing

2021 Kleinheubach Conference（2021）

引用 1|浏览9

暂无评分

摘要

A feasibility study of an energy efficient analog mixed-signal multiply-accumulate circuit with multi-bit resolution is presented. With a derived hardware-specific neuron model a ResNetV2 with 4 bit quantized weights is trained on a CIFAR10 dataset achieving 90% accuracy. The simulated energy efficiency of the multiply-accumulate circuit is 2.840 fJ per multiply-accumulate operation.

查看译文

关键词

mixed-signal computing,neural network inference,deep learning

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要