scispace - formally typeset
Open AccessPosted Content

Neural Architecture Search with Reinforcement Learning

Barret Zoph, +1 more
- 05 Nov 2016 - 
TLDR
This paper uses a recurrent network to generate the model descriptions of neural networks and trains this RNN with reinforcement learning to maximize the expected accuracy of the generated architectures on a validation set.
Abstract
Neural networks are powerful and flexible models that work well for many difficult learning tasks in image, speech and natural language understanding. Despite their success, neural networks are still hard to design. In this paper, we use a recurrent network to generate the model descriptions of neural networks and train this RNN with reinforcement learning to maximize the expected accuracy of the generated architectures on a validation set. On the CIFAR-10 dataset, our method, starting from scratch, can design a novel network architecture that rivals the best human-invented architecture in terms of test set accuracy. Our CIFAR-10 model achieves a test error rate of 3.65, which is 0.09 percent better and 1.05x faster than the previous state-of-the-art model that used a similar architectural scheme. On the Penn Treebank dataset, our model can compose a novel recurrent cell that outperforms the widely-used LSTM cell, and other state-of-the-art baselines. Our cell achieves a test set perplexity of 62.4 on the Penn Treebank, which is 3.6 perplexity better than the previous state-of-the-art model. The cell can also be transferred to the character language modeling task on PTB and achieves a state-of-the-art perplexity of 1.214.

read more

Citations
More filters
Proceedings ArticleDOI

AANG: Automating Auxiliary Learning

TL;DR: This work presents an approach for automatically generating a suite of auxiliary objectives by deconstructing existing objectives within a novel unified taxonomy, identifying connections between them, and generating new ones based on the uncovered structure and theoretically formalizes widely-held intuitions about how auxiliary learning improves generalization of the end-task.
Journal ArticleDOI

Automatic Development of Deep Learning Architectures for Image Segmentation

TL;DR: This work proposes an automatic method for creating new convolutional neural network architectures which are specifically designed to solve a given problem, and describes its reduced carbon footprint, computation time and cost compared to a manual approach.
Book ChapterDOI

A Committee of Convolutional Neural Networks for Image Classification in the Concurrent Presence of Feature and Label Noise

TL;DR: In this article, the authors propose three committee selection algorithms that outperform a strong baseline algorithm which relies on an ensemble of individual (nonassociated) best models, and experimentally prove that the difference by which committees beat single models increases along with noise level, no matter whether it is an attribute or label disruption.
Proceedings ArticleDOI

Basic Binary Convolution Unit for Binarized Image Restoration Network

TL;DR: This study reconsider components in binary convolution, such as residual connection, BatchNorm, activation function, and structure, for IR tasks, and designs a simple yetcient basic binary Convolution unit (BBCU), which shows that BBCU can serve as a basic unit for binarized IR networks.
Proceedings ArticleDOI

A Reinforcement Learning Based Design of Compressive Sensing Systems for Human Activity Recognition

TL;DR: Experimental results show that the the proposed system can increase the sensing efficiency and improve the performance of activity recognition.
References
More filters
Proceedings ArticleDOI

Deep Residual Learning for Image Recognition

TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Proceedings Article

Adam: A Method for Stochastic Optimization

TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Proceedings Article

Very Deep Convolutional Networks for Large-Scale Image Recognition

TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Journal ArticleDOI

Gradient-based learning applied to document recognition

TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Proceedings ArticleDOI

Histograms of oriented gradients for human detection

TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.