# Learning Time/Memory-Efficient Deep Architectures with Budgeted Super Networks

## TL;DR

This paper introduces Budgeted Super Networks, a novel approach for discovering neural network architectures optimized for both prediction accuracy and computational or memory costs, demonstrated on vision tasks.

## Contribution

The paper presents a new family of models called Budgeted Super Networks learned via gradient descent on a cost-aware objective, applicable to various resource constraints.

## Key findings

- BSN can discover architectures with better accuracy than ResNet and CNN Fabrics.
- BSN effectively handles different costs: computation, memory, and distributed computing.
- Models trained with BSN outperform existing architectures under similar resource budgets.

## Abstract

We propose to focus on the problem of discovering neural network architectures efficient in terms of both prediction quality and cost. For instance, our approach is able to solve the following tasks: learn a neural network able to predict well in less than 100 milliseconds or learn an efficient model that fits in a 50 Mb memory. Our contribution is a novel family of models called Budgeted Super Networks (BSN). They are learned using gradient descent techniques applied on a budgeted learning objective function which integrates a maximum authorized cost, while making no assumption on the nature of this cost. We present a set of experiments on computer vision problems and analyze the ability of our technique to deal with three different costs: the computation cost, the memory consumption cost and a distributed computation cost. We particularly show that our model can discover neural network architectures that have a better accuracy than the ResNet and Convolutional Neural Fabrics architectures on CIFAR-10 and CIFAR-100, at a lower cost.