期刊
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
卷 68, 期 1, 页码 138-147出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCSI.2020.3031627
关键词
Neural networks; resistive memory; quantized neural networks; low voltage operation; sense amplifier
资金
- ERC Grant NANOINFER [715872]
- ANR Grant NEURONIC [ANR-18-CE24-0009]
The paper focuses on the implementation of ternary neural networks and proposes a two-transistor/two-resistor memory architecture for achieving high energy efficiency at low supply voltage, while studying the bit error rate. Experimental results demonstrate that ternary neural networks can significantly improve neural network performance and exhibit immunity to bit errors.
The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a significant lead for reducing the energy consumption of artificial intelligence. To achieve maximum energy efficiency in such systems, logic and memory should be integrated as tightly as possible. In this work, we focus on the case of ternary neural networks, where synaptic weights assume ternary values. We propose a two-transistor/two-resistor memory architecture employing a precharge sense amplifier, where the weight value can be extracted in a single sense operation. Based on experimental measurements on a hybrid 130 nm CMOS/RRAM chip featuring this sense amplifier, we show that this technique is particularly appropriate at low supply voltage, and that it is resilient to process, voltage, and temperature variations. We characterize the bit error rate in our scheme. We show based on neural network simulation on the CIFAR-10 image recognition task that the use of ternary neural networks significantly increases neural network performance, with regards to binary ones, which are often preferred for inference hardware. We finally evidence that the neural network is immune to the type of bit errors observed in our scheme, which can therefore be used without error correction.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据