# AttoNets: Compact and Efficient Deep Neural Networks for the Edge via Human-Machine Collaborative Design

## TL;DR

This paper introduces AttoNets, a family of highly efficient deep neural networks designed for edge devices, created through a collaborative approach combining human insights and machine-driven design, resulting in models that are smaller, faster, and more accurate.

## Contribution

The paper presents a novel human-machine collaborative design methodology for creating efficient neural networks, exemplified by AttoNets, which outperform existing models in size, speed, and accuracy.

## Key findings

- AttoNets achieve ~1.8% higher accuracy than MobileNet-V1 with 10x fewer parameters.
- AttoNets require ~5x fewer multiply-add operations for object detection.
- AttoNets enable smaller, faster models for edge deep learning applications.

## Abstract

While deep neural networks have achieved state-of-the-art performance across a large number of complex tasks, it remains a big challenge to deploy such networks for practical, on-device edge scenarios such as on mobile devices, consumer devices, drones, and vehicles. In this study, we take a deeper exploration into a human-machine collaborative design approach for creating highly efficient deep neural networks through a synergy between principled network design prototyping and machine-driven design exploration. The efficacy of human-machine collaborative design is demonstrated through the creation of AttoNets, a family of highly efficient deep neural networks for on-device edge deep learning. Each AttoNet possesses a human-specified network-level macro-architecture comprising of custom modules with unique machine-designed module-level macro-architecture and micro-architecture designs, all driven by human-specified design requirements. Experimental results for the task of object recognition showed that the AttoNets created via human-machine collaborative design has significantly fewer parameters and computational costs than state-of-the-art networks designed for efficiency while achieving noticeably higher accuracy (with the smallest AttoNet achieving ~1.8% higher accuracy while requiring ~10x fewer multiply-add operations and parameters than MobileNet-V1). Furthermore, the efficacy of the AttoNets is demonstrated for the task of instance-level object segmentation and object detection, where an AttoNet-based Mask R-CNN network was constructed with significantly fewer parameters and computational costs (~5x fewer multiply-add operations and ~2x fewer parameters) than a ResNet-50 based Mask R-CNN network.