Posted Content•
Neural Architecture Search with Reinforcement 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
27 Sep 2022
TL;DR:
Abstract: Efficient and automated design of optimizers plays a crucial role in full-stack AutoML systems. However, prior methods in optimizer search are often limited by their scalability, generability, or sample efficiency. With the goal of democratizing research and application of optimizer search, we present the first efficient, scalable and generalizable framework that can directly search on the tasks of interest. We first observe that optimizer updates are fundamentally mathematical expressions applied to the gradient. Inspired by the innate tree structure of the underlying math expressions, we re-arrange the space of optimizers into a super-tree, where each path encodes an optimizer. This way, optimizer search can be naturally formulated as a path-finding problem, allowing a variety of well-established tree traversal methods to be used as the search algorithm. We adopt an adaptation of the Monte Carlo method to tree search, equipped with rejection sampling and equivalent-form detection that leverage the characteristics of optimizer update rules to further boost the sample efficiency. We provide a diverse set of tasks to benchmark our algorithm and demonstrate that, with only 128 evaluations, the proposed framework can discover optimizers that surpass both human-designed counterparts and prior optimizer search methods.
3 citations
TL;DR: PENGUIN is a decoupled fitness prediction engine that informs the search without interfering in it, and reduces training costs: it predicts the fitness of NNs, enabling NAS to terminate training NNs early.
Abstract: Neural networks (NN) are used in high-performance computing and high-throughput analysis to extract knowledge from datasets. Neural architecture search (NAS) automates NN design by generating, training, and analyzing thousands of NNs. However, NAS requires massive computational power for NN training. To address challenges of efficiency and scalability, we propose PENGUIN, a decoupled fitness prediction engine that informs the search without interfering in it. PENGUIN uses parametric modeling to predict fitness of NNs. Existing NAS methods and parametric modeling functions can be plugged into PENGUIN to build flexible NAS workflows. Through this decoupling and flexible parametric modeling, PENGUIN reduces training costs: it predicts the fitness of NNs, enabling NAS to terminate training NNs early. Early termination increases the number of NNs that fixed compute resources can evaluate, thus giving NAS additional opportunity to find better NNs. We assess the effectiveness of our engine on 6,000 NNs across three diverse benchmark datasets and three state of the art NAS implementations using the Summit supercomputer. Augmenting these NAS implementations with PENGUIN can increase throughput by a factor of 1.6 to 7.1.Furthermore, walltime tests indicate that PENGUIN can reduce training time by a factor of 2.5 to 5.3.
3 citations
TL;DR: It is proved that the optimal power allocation policy and the optimal action-value function depend monotonically on some of their input variables and the shallow neural network structure is designed based on properties revealed in the proof.
Abstract: This article considers a power allocation problem in energy harvesting downlink non-orthogonal multiple access (NOMA) systems in which a transmitter sends desired messages to their respective receivers by using harvested energy. To tackle this problem, we make use of a reinforcement learning approach based on a shallow neural network structure. We prove that the optimal power allocation policy and the optimal action-value function depend monotonically on some of their input variables and the shallow neural network structure is designed based on properties revealed in the proof. Different from inefficient deep learning methods that tend to require tremendous computational resources, this structure is capable of fully capturing the characteristics of the desired function with a single hidden layer. The optimized structure also allows learning agents to be robust and highly reliable in learning about randomly occurring data. Furthermore, we provide comprehensive experimental results in harsh environments where various arbitrary factors are assumed in order to demonstrate the robustness of the proposed learning approach compared with deep neural networks without proper grounds. It is also shown that the proposed learning process converges to a policy that outperforms existing power allocation algorithms.
3 citations
Cites background from "Neural Architecture Search with Rei..."
...In order to optimize the network for specific problems, huge learning structures that learns the network structure itself have been proposed in [41]....
[...]
TL;DR: A reinforcement transfer learning method based on a policy gradient to identify the optimal structure of an intelligent fault diagnosis model when the number of training samples is insufficient, confirming the importance of the autonomous selection of a diagnostic model structure.
Abstract: Accurate rolling bearing fault diagnosis is a basic guarantee of the safe operation of rotating machinery. Therefore, it is critical to select an appropriate fault diagnosis model. Selecting the optimal structure for intelligent fault diagnosis model has gradually become a research hot spot. At the same time, in practical engineering, sufficient data cannot always be guaranteed, which also increases the difficulty of accurate fault diagnosis. This paper proposes a reinforcement transfer learning method based on a policy gradient to identify the optimal structure of an intelligent fault diagnosis model when the number of training samples is insufficient. First, a policy gradient method is used to select the optimal child model in the source domain. Second, a transfer learning method is adopted to transfer the hyperparameters of the optimal child model from the source domain to the target domain. Finally, a small number of labeled training samples are used to fine-tune this model in the target domain. An adequate number of experiments proved the viability of proposed method, confirming the importance of the autonomous selection of a diagnostic model structure.
3 citations
TL;DR: An approach is designed to search the segmentation loss function based on the computation graphs according to the evaluation metrics, and the evolution algorithm is revised to avoid potential loss and equivalent loss functions.
3 citations
References
More filters
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 2015TL;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 2014TL;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)....
[...]
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
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)....
[...]