Multi-Objective Neural Architecture Search for In-Memory Computing

TL;DR

This paper applies neural architecture search with Bayesian optimization to design efficient CNN models tailored for in-memory computing architectures, optimizing for accuracy, latency, and energy across multiple datasets.

Contribution

It introduces a novel NAS framework for IMC architectures, exploring over 640 million configurations to find balanced, high-performance neural networks.

Findings

Achieved high accuracy with reduced latency and energy consumption.

Demonstrated effectiveness across three image classification datasets.

Optimized multi-objective cost functions successfully.

Abstract

In this work, we employ neural architecture search (NAS) to enhance the efficiency of deploying diverse machine learning (ML) tasks on in-memory computing (IMC) architectures. Initially, we design three fundamental components inspired by the convolutional layers found in VGG and ResNet models. Subsequently, we utilize Bayesian optimization to construct a convolutional neural network (CNN) model with adaptable depths, employing these components. Through the Bayesian search algorithm, we explore a vast search space comprising over 640 million network configurations to identify the optimal solution, considering various multi-objective cost functions like accuracy/latency and accuracy/energy. Our evaluation of this NAS approach for IMC architecture deployment spans three distinct image classification datasets, demonstrating the effectiveness of our method in achieving a balanced solution characterized by high accuracy and reduced latency and energy consumption.