Litcius/Paper detail

An Energy Efficient Computing-in-Memory Accelerator With 1T2R Cell and Fully Analog Processing for Edge AI Applications

Keji Zhou, Chenyang Zhao, Jinbei Fang, Jingwen Jiang, Deyang Chen, Yujie Huang, Minge Jing, Jun Han, Haidong Tian, Xiankui Xiong, Qi Liu, Xiaoyong Xue, Xiaoyang Zeng

2021IEEE Transactions on Circuits & Systems II Express Briefs22 citationsDOI

Abstract

In this work, a ReRAM-based energy-efficient CIM accelerator is presented with two techniques for edge AI applications. Firstly, a circuit-algorithm co-design scheme is proposed to realize fully analog processing, which improves the energy efficiency and the throughput of neural network. To deal with the I-V nonlinearity of ReRAM, a nonlinear-aware training algorithm is proposed to improve the network accuracy. Secondly, a 1T2R cell is proposed to replace previous 2T2R for weight storage with 35% area saving. For evaluation, a neural network with two fully connected layers and one ReLU layer is built for the MNIST dataset. The error rate can be reduced by >46% and the energy efficiency is 99 TOPS/W@200 MHz, 2.6X improvement over the digital method.

Topics & Concepts

MNIST databaseComputer scienceArtificial neural networkEfficient energy useEnergy (signal processing)Enhanced Data Rates for GSM EvolutionThroughputEdge deviceResistive random-access memoryComputer hardwareNonlinear systemElectronic engineeringComputer engineeringArtificial intelligenceElectrical engineeringVoltageEngineeringTelecommunicationsMathematicsQuantum mechanicsWirelessPhysicsStatisticsCloud computingOperating systemAdvanced Memory and Neural ComputingFerroelectric and Negative Capacitance DevicesNeural Networks and Reservoir Computing