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
01 Dec 2019
TL;DR: An automated device-level DNN model selection engine for QoE-optimal edge inference by leveraging the predictive power of machine learning and keeping end users in the loop is envisioned.
Abstract: The ever increasing size of deep neural network (DNN) models once implied that they were only limited to cloud data centers for runtime inference. Nonetheless, the recent plethora of DNN model compression techniques have successfully overcome this limit, turning into a reality that DNN-based inference can be run on numerous resource-constrained edge devices including mobile phones, drones, robots, medical devices, wearables, Internet of Things devices, among many others. Naturally, edge devices are highly heterogeneous in terms of hardware specification and usage scenarios. On the other hand, compressed DNN models are so diverse that they exhibit different tradeoffs in a multi-dimension space, and not a single model can achieve optimality in terms of all important metrics such as accuracy, latency and energy consumption. Consequently, how to automatically select a compressed DNN model for an edge device to run inference with optimal quality of experience (QoE) arises as a new challenge. The state-of-the-art approaches either choose a common model for all/most devices, which is optimal for a small fraction of edge devices at best, or apply device-specific DNN model compression, which is not scalable. In this paper, by leveraging the predictive power of machine learning and keeping end users in the loop, we envision an automated device-level DNN model selection engine for QoE-optimal edge inference. To concretize our vision, we formulate the DNN model selection problem into a contextual multi-armed bandit framework, where features of edge devices and DNN models are contexts and pre-trained DNN models are arms selected online based on the history of actions and users' QoE feedback. We develop an efficient online learning algorithm to balance exploration and exploitation. Our preliminary simulation results validate our algorithm and highlight the potential of machine learning for automating DNN model selection to achieve QoE-optimal edge inference.
9 citations
Cites background or methods from "Neural Architecture Search with Rei..."
...Last but not least, even with the same training data set but given different compression techniques or optimization parameters, hundreds of or even more different DNN models can be generated using neural architecture search techniques [9], [10], among which no single model is a clear winner in terms of all important metrics such as accuracy, latency and energy....
[...]
...As a consequence, even with the same training data set but given different compression techniques or optimization parameters, hundreds of or even more different DNN models can be generated using neural architecture search techniques [9], [10]....
[...]
...On the other hands, Oracle represents a “one for one” approach (similar to those proposed by recent studies [4], [5], [9], [10], [13], [15]–[17]) and chooses the best DNN model for each individual edge device to maximize the expected QoE, under the assumption that the parameter θ in the QoE model is perfectly known a priori....
[...]
...2) Poor scalability: Many research studies have leveraged AutoML and neural architecture search techniques to identify the optimal DNN model for a specific target edge device [4], [5], [9], [10], [13], [15]–[17]....
[...]
...Given a large set of DNN models pre-trained in the cloud using state-of-the-art AutoML and neural architecture search techniques [9], [10], [13], [15], our DNN model selection engine constructs a QoE model based on users’ QoE feedback and selects appropriate DNN models for edge devices, balancing exploration and exploitation....
[...]
Posted Content•
TL;DR: This work proposes an evolutionary algorithm for searching neural architectures under multiple objectives, such as classification performance and FLOPs, and addresses the first shortcoming by populating a set of architectures to approximate the entire Pareto frontier through genetic operations that recombine and modify architectural components progressively.
Abstract: Convolutional neural networks (CNNs) are the backbones of deep learning paradigms for numerous vision tasks. Early advancements in CNN architectures are primarily driven by human expertise and elaborate design. Recently, neural architecture search was proposed with the aim of automating the network design process and generating task-dependent architectures. While existing approaches have achieved competitive performance in image classification, they are not well suited under limited computational budget for two reasons: (1) the obtained architectures are either solely optimized for classification performance or only for one targeted resource requirement; (2) the search process requires vast computational resources in most approaches. To overcome this limitation, we propose an evolutionary algorithm for searching neural architectures under multiple objectives, such as classification performance and FLOPs. The proposed method addresses the first shortcoming by populating a set of architectures to approximate the entire Pareto frontier through genetic operations that recombine and modify architectural components progressively. Our approach improves the computation efficiency by carefully down-scaling the architectures during the search as well as reinforcing the patterns commonly shared among the past successful architectures through Bayesian Learning. The integration of these two main contributions allows an efficient design of architectures that are competitive and in many cases outperform both manually and automatically designed architectures on benchmark image classification datasets, CIFAR, ImageNet and human chest X-ray. The flexibility provided from simultaneously obtaining multiple architecture choices for different compute requirements further differentiates our approach from other methods in the literature.
9 citations
Cites background or methods from "Neural Architecture Search with Rei..."
...The original method in [8] uses the controller to generate the entire network at once....
[...]
...A major limitation of these approaches [8], [9] is the steep computational requirement for the search process itself, often requiring weeks of wall clock time on hundreds of GPU cards....
[...]
...The second and third groups range from earlier methods [8], [16], [17] that are oriented towards “proof-of-concept" for NAS, to more recent methods [9], [10], [15], [19], many of which improve state-of-the-art results on...
[...]
...Zoph and Le [8] first apply this method in architecture search to train a recurrent neural network controller that constructs networks....
[...]
Posted Content•
TL;DR: This work proposes an adapted implementation of a well-established evolutionary technique from the neuroevolution field that manages to automate the tasks of topology and hyperparameter selection.
Abstract: Machine learning is a huge field of study in computer science and statistics dedicated to the execution of computational tasks through algorithms that do not require explicit instructions but instead rely on learning patterns from data samples to automate inferences. A large portion of the work involved in a machine learning project is to define the best type of algorithm to solve a given problem. Neural networks - especially deep neural networks - are the predominant type of solution in the field. However, the networks themselves can produce very different results according to the architectural choices made for them. Finding the optimal network topology and configurations for a given problem is a challenge that requires domain knowledge and testing efforts due to a large number of parameters that need to be considered. The purpose of this work is to propose an adapted implementation of a well-established evolutionary technique from the neuroevolution field that manages to automate the tasks of topology and hyperparameter selection. It uses a popular and accessible machine learning framework - Keras - as the back-end, presenting results and proposed changes concerning the original algorithm. The implementation is available at GitHub (this https URL) with documentation and examples to reproduce the experiments performed for this work.
9 citations
Additional excerpts
...This rapid popularization of different types of neural networks brought into the neuroevolution field challenges to try new techniques by combining and expanding these varied components into appropriate topologies and configurations to solve problems even more effectively and is also referred to as the neural architecture search problem [12]....
[...]
01 Nov 2020
TL;DR: A multi-task shared structure encoding approach which automatically selects good shared network structures as well as good auxiliary resources for improving the performance of the target.
Abstract: Automatic prediction on the peer-review aspect scores of academic papers can be a useful assistant tool for both reviewers and authors. To handle the small size of published datasets on the target aspect of scores, we propose a multi-task approach to leverage additional information from other aspects of scores for improving the performance of the target. Because one of the problems of building multi-task models is how to select the proper resources of auxiliary tasks and how to select the proper shared structures. We propose a multi-task shared structure encoding approach which automatically selects good shared network structures as well as good auxiliary resources. The experiments based on peer-review datasets show that our approach is effective and has better performance on the target scores than the single-task method and naive multi-task methods.
9 citations
Cites background from "Neural Architecture Search with Rei..."
..., 2012); the network architecture search (NAS) methods based on reinforcement learning (Zoph and Le, 2016; Zoph et al., 2018; Liu et al., 2018)....
[...]
..., 2012) or network architecture search methods (Zoph and Le, 2016) for searching....
[...]
03 Nov 2022
TL;DR: In this article , a general-purpose technique that selectively performs computation for edited regions and accelerates various generative models, including both conditional GANs and diffusion models, is presented.
Abstract: During image editing, existing deep generative models tend to re-synthesize the entire output from scratch, including the unedited regions. This leads to a significant waste of computation, especially for minor editing operations. In this work, we present Spatially Sparse Inference (SSI), a general-purpose technique that selectively performs computation for edited regions and accelerates various generative models, including both conditional GANs and diffusion models. Our key observation is that users tend to gradually change the input image. This motivates us to cache and reuse the feature maps of the original image. Given an edited image, we sparsely apply the convolutional filters to the edited regions while reusing the cached features for the unedited areas. Based on our algorithm, we further propose Sparse Incremental Generative Engine (SIGE) to convert the computation reduction to latency reduction on off-the-shelf hardware. With about $1\%$-area edits, our method reduces the computation of DDPM by $7.5\times$, Stable Diffusion by $8.2\times$, and GauGAN by $18\times$ while preserving the visual fidelity. With SIGE, we accelerate the inference time of DDPM by $3.0\times$ on NVIDIA RTX 3090 and $6.6\times$ on Apple M1 Pro CPU, Stable Diffusion by $7.2\times$ on 3090, and GauGAN by $5.6\times$ on 3090 and $14\times$ on M1 Pro CPU.
9 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)....
[...]