We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

There are a huge number of features which are said to improve Convolutional Neural Network (CNN) accuracy. Practical testing of combinations of such features on large datasets, and theoretical justification of the result, is required. Some features operate on certain models exclusively and for certain problems exclusively, or only for small-scale datasets; while some features, such as batch-normalization and residual-connections, are applicable to the majority of models, tasks, and datasets. We assume that such universal features include Weighted-Residual-Connections (WRC), Cross-Stage-Partial-connections (CSP), Cross mini-Batch Normalization (CmBN), Self-adversarial-training (SAT) and Mish-activation. We use new features: WRC, CSP, CmBN, SAT, Mish activation, Mosaic data augmentation, CmBN, DropBlock regularization, and CIoU loss, and combine some of them to achieve state-of-the-art results: 43.5% AP (65.7% AP50) for the MS COCO dataset at a realtime speed of ~65 FPS on Tesla V100. Source code is at this https URL

YOLOv4: Optimal Speed and Accuracy of Object Detection

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Image animation is to animate a still image of the object of interest using poses extracted from another video sequence. Through training on a large-scale video dataset, most existing approaches aim to explore disentangled appearance and pose representations of training frames. Then, the desired output with a specific appearance and pose can be synthesized via recombining learned representations. However, in some real-world applications, test images may lack the corresponding video ground-truth or follow a different distribution than the distribution of the training video frames (i.e., different domains), which largely limit the performance of existing methods. In this paper, we propose domain-independent pose representations that are compatible with and accessible by still images from a different domain. Specifically, we devise a two-stage self-supervised pose adaptation framework for general image animation tasks. A domain-independent pose adaptation generative adversarial network (DIPA-GAN) and a shuffle-patch generative adversarial network (Shuffle-patch GAN) are proposed to penalize the rationality of the synthesized frame's pose and appearance, respectively. Finally, experiments evaluated on various image animation tasks, which include same/cross-domain moving objects, facial expression transfer and human pose retargeting, demonstrate the superiority of the proposed framework over prior literature. Impact Statement —Image animation is a popular technology in video production. Benefiting from the rapid development of artificial intelligence (AI), recent image animation algorithms have been widely used in real-world applications, such as virtual AI news anchor, virtual try-on, and face swapping. However, most existing methods are designed for specific cases. To animate a new portrait, users are asked to collect hundreds of images of the same person and train a new model. The technology proposed in this paper overcomes these training limitations and generalizes image animations. In the challenging cross-domain facial expression transfer task, the user study demonstrated that our technology achieved more than 20% increase in animation success rate. The proposed technology could benefit users in a wide variety of industries including movie production, virtual reality, social media and online retail.

Self-Supervised Pose Adaptation for Cross-Domain Image Animation

Neural architecture search (NAS) has demonstrated impressive performance in automatically designing high-performance neural networks. The power of deep neural networks is to be unleashed for analyzing a large volume of data (e.g. ImageNet), but the architecture search is often executed on another smaller dataset (e.g. CIFAR-10) to finish it in a feasible time. However, it is hard to guarantee that the optimal architecture derived on the proxy task could maintain its advantages on another more challenging dataset. This paper aims to improve the generalization of neural architectures via domain adaptation. We analyze the generalization bounds of the derived architecture and suggest its close relations with the validation error and the data distribution distance on both domains. These theoretical analyses lead to AdaptNAS, a novel and principled approach to adapt neural architectures between domains in NAS. Our experimental evaluation shows that only a small part of ImageNet will be sufficient for AdaptNAS to extend its architecture success to the entire ImageNet and outperform state-of-the-art comparison algorithms.

/pdf/adapting-neural-architectures-between-domains-1p386rg673.pdf

Adapting Neural Architectures Between Domains

Data-free learning for student networks is a new paradigm for solving users’ anxiety caused by the privacy problem of using original training data. Since the architectures of modern convolutional neural networks (CNNs) are compact and sophisticated, the alternative images or meta-data generated from the teacher network are often broken. Thus, the student network cannot achieve the comparable performance to that of the pre-trained teacher network especially on the large-scale image dataset. Different to previous works, we present to maximally utilize the massive available unlabeled data in the wild. Specifically, we first thoroughly analyze the output differences between teacher and student network on the original data and develop a data collection method. Then, a noisy knowledge distillation algorithm is proposed for achieving the performance of the student network. In practice, an adaptation matrix is learned with the student network for correcting the label noise produced by the teacher network on the collected unlabeled images. The effectiveness of our DFND (Data-Free Noisy Distillation) method is then verified on several benchmarks to demonstrate its superiority over state-of-the-art data-free distillation methods. Experiments on various datasets demonstrate that the student networks learned by the proposed method can achieve comparable performance with those using the original dataset. Code is available at https://github.com/huawei-noah/Data-Efficient-Model-Compression

/pdf/learning-student-networks-in-the-wild-1xvni2h2vr.pdf

Learning Student Networks in the Wild

Neural Architecture Search (NAS) refers to automatically design the architecture. We propose an hourglass-inspired approach (HourNAS) for this problem that is motivated by the fact that the effects of the architecture often proceed from the vital few blocks. Acting like the narrow neck of an hourglass, vital blocks in the guaranteed path from the input to the output of a deep neural network restrict the information flow and influence the network accuracy. The other blocks occupy the major volume of the network and determine the overall network complexity, corresponding to the bulbs of an hourglass. To achieve an extremely fast NAS while preserving the high accuracy, we propose to identify the vital blocks and make them the priority in the architecture search. The search space of those non-vital blocks is further shrunk to only cover the candidates that are affordable under the computational resource constraints. Experimental results on the ImageNet show that only using 3 hours (0.1 days) with one GPU, our HourNAS can search an architecture that achieves a 77.0% Top-1 accuracy, which outperforms the state-of-the-art methods.

HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass Lens

Compressing convolutional neural networks (CNNs) is essential for transferring the success of CNNs to a wide variety of applications to mobile devices. In contrast to directly recognizing subtle weights or filters as redundant in a given CNN, this paper presents an evolutionary method to automatically eliminate redundant convolution filters. We represent each compressed network as a binary individual of specific fitness. Then, the population is upgraded at each evolutionary iteration using genetic operations. As a result, an extremely compact CNN is generated using the fittest individual. In this approach, either large or small convolution filters can be redundant, and filters in the compressed network are more distinct. In addition, since the number of filters in each convolutional layer is reduced, the number of filter channels and the size of feature maps are also decreased, naturally improving both the compression and speed-up ratios. Experiments on benchmark deep CNN models suggest the superiority of the proposed algorithm over the state-of-the-art compression methods.

Chang Xu

Papers

Self-Supervised Pose Adaptation for Cross-Domain Image Animation

Adapting Neural Architectures Between Domains

Learning Student Networks in the Wild

HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass Lens

Towards Evolutional Compression