scispace - formally typeset
Search or ask a question
Posted Content

Neural Architecture Search with Reinforcement Learning

Barret Zoph1, Quoc V. Le1
05 Nov 2016-arXiv: Learning-
TL;DR: 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.
Citations
More filters
Proceedings ArticleDOI
01 Nov 2019
TL;DR: This paper improves differentiable architecture search by removing the softmax-local constraint and applies differentiable NAS to named entity recognition (NER), the first time that differentiableNAS methods are adopted in NLP tasks other than language modeling.
Abstract: In this paper, we study differentiable neural architecture search (NAS) methods for natural language processing. In particular, we improve differentiable architecture search by removing the softmax-local constraint. Also, we apply differentiable NAS to named entity recognition (NER). It is the first time that differentiable NAS methods are adopted in NLP tasks other than language modeling. On both the PTB language modeling and CoNLL-2003 English NER data, our method outperforms strong baselines. It achieves a new state-of-the-art on the NER task.

84 citations


Cites background or methods from "Neural Architecture Search with Rei..."

  • ...The earliest of these approaches use reinforcement learning (RL) to learn promising architectures in a discrete space (Zoph and Le, 2016), whereas others have successfully modeled the problem in a continuous manner (Liu et al....

    [...]

  • ...Like recent methods in NAS (Xie and Yuille, 2017; Zoph and Le, 2016; Baker et al., 2016), DARTS represents networks as a directed acyclic graph for a given computation cell (see Figure 1(a))....

    [...]

  • ...Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 3576–3581, Hong Kong, China, November 3–7, 2019. c©2019 Association for Computational Linguistics 3576 In this paper, we study differentiable neural architecture search (NAS) methods for natural language processing....

    [...]

  • ...It is the first time that differentiable NAS methods are adopted in NLP tasks other than language modeling....

    [...]

  • ...Also, we apply differentiable NAS to named entity recognition (NER)....

    [...]

Posted Content
TL;DR: This paper explores effective Transformer architectures for language model, including adding additional LSTM layers to better capture the sequential context while still keeping the computation efficient, and proposes Coordinate Architecture Search (CAS) to find an effective architecture through iterative refinement of the model.
Abstract: The Transformer architecture is superior to RNN-based models in computational efficiency. Recently, GPT and BERT demonstrate the efficacy of Transformer models on various NLP tasks using pre-trained language models on large-scale corpora. Surprisingly, these Transformer architectures are suboptimal for language model itself. Neither self-attention nor the positional encoding in the Transformer is able to efficiently incorporate the word-level sequential context crucial to language modeling. In this paper, we explore effective Transformer architectures for language model, including adding additional LSTM layers to better capture the sequential context while still keeping the computation efficient. We propose Coordinate Architecture Search (CAS) to find an effective architecture through iterative refinement of the model. Experimental results on the PTB, WikiText-2, and WikiText-103 show that CAS achieves perplexities between 20.42 and 34.11 on all problems, i.e. on average an improvement of 12.0 perplexity units compared to state-of-the-art LSTMs. The source code is publicly available.

84 citations


Cites background or methods from "Neural Architecture Search with Rei..."

  • ...The GPT vocabulary size is 10k larger compared to the ENAS’s vocabulary....

    [...]

  • ...Due to its greedy nature, CAS is simpler and faster than previous architecture search algorithms (Zoph and Le, 2016; Pham et al., 2018; Liu et al., 2018)....

    [...]

  • ..., 2018) as well as language modeling (Zoph and Le, 2016; Pham et al., 2018; Liu et al., 2018) in NLP....

    [...]

  • ...The reason of the search cost of GPT-CAS on WT-2 is higher than ENAS is three-fold: 1....

    [...]

  • ...ENAS by (Pham et al., 2018) also leverages a reinforcement learning search method....

    [...]

Posted Content
TL;DR: It is shown that DEQs often improve performance over these state-of-the-art models (for similar parameter counts); have similar computational requirements to existing models; and vastly reduce memory consumption (often the bottleneck for training large sequence models), demonstrating an up-to 88% memory reduction in the authors' experiments.
Abstract: We present a new approach to modeling sequential data: the deep equilibrium model (DEQ). Motivated by an observation that the hidden layers of many existing deep sequence models converge towards some fixed point, we propose the DEQ approach that directly finds these equilibrium points via root-finding. Such a method is equivalent to running an infinite depth (weight-tied) feedforward network, but has the notable advantage that we can analytically backpropagate through the equilibrium point using implicit differentiation. Using this approach, training and prediction in these networks require only constant memory, regardless of the effective "depth" of the network. We demonstrate how DEQs can be applied to two state-of-the-art deep sequence models: self-attention transformers and trellis networks. On large-scale language modeling tasks, such as the WikiText-103 benchmark, we show that DEQs 1) often improve performance over these state-of-the-art models (for similar parameter counts); 2) have similar computational requirements to existing models; and 3) vastly reduce memory consumption (often the bottleneck for training large sequence models), demonstrating an up-to 88% memory reduction in our experiments. The code is available at this https URL .

84 citations

Proceedings Article
04 Nov 2018
TL;DR: A new data augmentation technique called random image cropping and patching (RICAP), which randomly crops four images and patches them to construct a new training image, enriching the variety of training images.
Abstract: Deep convolutional neural networks (CNNs) have demonstrated remarkable results in image recognition owing to their rich expression ability and numerous parameters. However, an excessive expression ability compared to the variety of training images often has a risk of overfitting. Data augmentation techniques have been proposed to address this problem as they enrich datasets by flipping, cropping, resizing, and color-translating images. They enable deep CNNs to achieve an impressive performance. In this study, we propose a new data augmentation technique called random image cropping and patching (RICAP), which randomly crops four images and patches them to construct a new training image. Hence, RICAP randomly picks up subsets of original features among the four images and discard others, enriching the variety of training images. Also, RICAP mixes the class labels of the four images and enjoys a benefit similar to label smoothing. We evaluated RICAP with current state-of-the-art CNNs (e.g., shake-shake regularization model) and achieved a new state-of-the-art test error of 2.23% on CIFAR-10 among competitive data augmentation techniques such as cutout and mixup. We also confirmed that deep CNNs with RICAP achieved better results on CIFAR-100 and ImageNet than those results obtained by other techniques.

84 citations

Posted Content
TL;DR: A unified way to encode individual layers into vectors and bring them together to form an integrated description via LSTM, taking advantage of the recurrent network's strong expressive power, can reliably predict the performances of various network architectures.
Abstract: The quest for performant networks has been a significant force that drives the advancements of deep learning in recent years. While rewarding, improving network design has never been an easy journey. The large design space combined with the tremendous cost required for network training poses a major obstacle to this endeavor. In this work, we propose a new approach to this problem, namely, predicting the performance of a network before training, based on its architecture. Specifically, we develop a unified way to encode individual layers into vectors and bring them together to form an integrated description via LSTM. Taking advantage of the recurrent network's strong expressive power, this method can reliably predict the performances of various network architectures. Our empirical studies showed that it not only achieved accurate predictions but also produced consistent rankings across datasets -- a key desideratum in performance prediction.

84 citations


Cites background or methods from "Neural Architecture Search with Rei..."

  • ...been reported [27, 28] that searching a network on CIFAR takes hundreds of GPUs for a lengthy period....

    [...]

  • ...Several methods [27, 1, 28, 26] have been proposed....

    [...]

  • ...In recent works [27, 1, 28, 26], automatic search methods have been proposed....

    [...]

References
More filters
Proceedings ArticleDOI
27 Jun 2016
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.
Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

123,388 citations

Proceedings Article
01 Jan 2015
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.
Abstract: We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments. The method is straightforward to implement, is computationally efficient, has little memory requirements, is invariant to diagonal rescaling of the gradients, and is well suited for problems that are large in terms of data and/or parameters. The method is also appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. Empirical results demonstrate that Adam works well in practice and compares favorably to other stochastic optimization methods. Finally, we discuss AdaMax, a variant of Adam based on the infinity norm.

111,197 citations

Proceedings Article
04 Sep 2014
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.
Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) 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. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

55,235 citations


"Neural Architecture Search with Rei..." refers methods in this paper

  • ...Along with this success is a paradigm shift from feature designing to architecture designing, i.e., from SIFT (Lowe, 1999), and HOG (Dalal & Triggs, 2005), to AlexNet (Krizhevsky et al., 2012), VGGNet (Simonyan & Zisserman, 2014), GoogleNet (Szegedy et al., 2015), and ResNet (He et al., 2016a)....

    [...]

Journal ArticleDOI
01 Jan 1998
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.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.

42,067 citations

Proceedings ArticleDOI
20 Jun 2005
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.
Abstract: We study the question of feature sets for robust visual object recognition; adopting linear SVM based human detection as a test case. After reviewing existing edge and gradient based descriptors, we show experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection. We study the influence of each stage of the computation on performance, concluding that fine-scale gradients, fine orientation binning, relatively coarse spatial binning, and high-quality local contrast normalization in overlapping descriptor blocks are all important for good results. The new approach gives near-perfect separation on the original MIT pedestrian database, so we introduce a more challenging dataset containing over 1800 annotated human images with a large range of pose variations and backgrounds.

31,952 citations


"Neural Architecture Search with Rei..." refers methods in this paper

  • ...Along with this success is a paradigm shift from feature designing to architecture designing, i.e., from SIFT (Lowe, 1999), and HOG (Dalal & Triggs, 2005), to AlexNet (Krizhevsky et al., 2012), VGGNet (Simonyan & Zisserman, 2014), GoogleNet (Szegedy et al., 2015), and ResNet (He et al., 2016a)....

    [...]