XPert: Peripheral Circuit Neural Architecture Co-search for Area and Energy-efficient Xbar-based Computing

03/30/2023

∙

The hardware-efficiency and accuracy of Deep Neural Networks (DNNs) implemented on In-memory Computing (IMC) architectures primarily depend on the DNN architecture and the peripheral circuit parameters. It is therefore essential to holistically co-search the network and peripheral parameters to achieve optimal performance. To this end, we propose XPert, which co-searches network architecture in tandem with peripheral parameters such as the type and precision of analog-to-digital converters, crossbar column sharing and the layer-specific input precision using an optimization-based design space exploration. Compared to VGG16 baselines, XPert achieves 10.24x (4.7x) lower EDAP, 1.72x (1.62x) higher TOPS/W,1.93x (3x) higher TOPS/mm2 at 92.46 accuracy for CIFAR10 (TinyImagenet) datasets. The code for this paper is available at https://github.com/Intelligent-Computing-Lab-Yale/XPert.

READ FULL TEXT

XPert: Peripheral Circuit Neural Architecture Co-search for Area and Energy-efficient Xbar-based Computing

NAX: Co-Designing Neural Network and Hardware Architecture for Memristive Xbar based Computing Systems

SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size

Evolving Neural Networks through a Reverse Encoding Tree

Using the IBM Analog In-Memory Hardware Acceleration Kit for Neural Network Training and Inference

HADAS: Hardware-Aware Dynamic Neural Architecture Search for Edge Performance Scaling

OSA-HCIM: On-The-Fly Saliency-Aware Hybrid SRAM CIM with Dynamic Precision Configuration

ERSAM: Neural Architecture Search For Energy-Efficient and Real-Time Social Ambiance Measurement

XPert: Peripheral Circuit Neural Architecture Co-search for Area and Energy-efficient Xbar-based Computing

Related Research

NAX: Co-Designing Neural Network and Hardware Architecture for Memristive Xbar based Computing Systems

SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size

Evolving Neural Networks through a Reverse Encoding Tree

Using the IBM Analog In-Memory Hardware Acceleration Kit for Neural Network Training and Inference

HADAS: Hardware-Aware Dynamic Neural Architecture Search for Edge Performance Scaling

OSA-HCIM: On-The-Fly Saliency-Aware Hybrid SRAM CIM with Dynamic Precision Configuration

ERSAM: Neural Architecture Search For Energy-Efficient and Real-Time Social Ambiance Measurement