Litcius/Paper detail

A Multiply-Less Approximate SRAM Compute-In-Memory Macro for Neural-Network Inference

Haikang Diao, Yifan He, Xuan Li, Chen Tang, Wenbin Jia, Jinshan Yue, Haoyang Luo, Jiahao Song, Xueqing Li, Huazhong Yang, Hongyang Jia, Yongpan Liu, Yuan Wang, Xiyuan Tang

2024IEEE Journal of Solid-State Circuits16 citationsDOI

Abstract

Compute-in-memory (CIM) is promising in reducing data movement energy and providing large bandwidth for matrix-vector multiplies (MVMs). However, existing work still faces various challenges, such as the digital logic overhead caused by the multiply-add operations (OPs) and structural sparsity. This article presents a 2-to-8-b scalable approximate digital SRAM-based CIM macro co-designed with a multiply-less neural network (NN) approach. It incorporates dynamic-logic-based approximate circuits for the logic area and energy saving by eliminating multiplications. A prototype is fabricated in 28-nm CMOS technology and achieves peak multiply-accumulate (MAC)-level energy efficiency of 102 TOPS/W for 8-b operations. The NN model deployment flow is used to demonstrate CIFAR-10 and ImageNet classification with ResNet-20 and ResNet-50 style multiply-less models, respectively, achieving the accuracy of 91.74% and 74.8% with 8-bit weights and activations.

Topics & Concepts

Static random-access memoryMacroInferenceComputer scienceArtificial neural networkAlgorithmParallel computingArtificial intelligenceProgramming languageComputer hardwareAdvanced Memory and Neural ComputingVLSI and Analog Circuit TestingNeural Networks and Applications