Unsupervised feature learning-based encoder and adversarial networks

AbstractIn this paper, we propose a novel deep learning-based feature learning architecture for object classification. Conventionally, deep learning methods are trained with supervised learning for object classification. But, this would require large amount of training data. Currently there are increasing trends to employ unsupervised learning for deep learning. By doing so, dependency on the availability of large training data could be reduced. One implementation of unsupervised deep learning is for feature learning where the network is designed to “learn” features automatically from data to obtain good representation that then could be used for classification. Autoencoder and generative adversarial networks (GAN) are examples of unsupervised deep learning methods. For GAN however, the trajectories of feature learning may go to unpredicted directions due to random initialization, making it unsuitable for feature learning. To overcome this, a hybrid of encoder and deep convolutional generative adversarial network (DCGAN) architectures, a variant of GAN, are proposed. Encoder is put on top of the Generator networks of GAN to avoid random initialisation. We called our method as EGAN. The output of EGAN is used as features for two deep convolutional neural networks (DCNNs): AlexNet and DenseNet. We evaluate the proposed methods on three types of dataset and the results indicate that better performances are achieved by our proposed method compared to using autoencoder and GAN.

Download Full-text

Improving Cloud-based ECG Monitoring, Detection and Classification using GAN

10.54216/fpa.020201 ◽

2020 ◽

pp. 42-49

Author(s):

admin admin ◽

◽

Monika Gupta

Keyword(s):

Deep Learning ◽

Training Data ◽

Generative Adversarial Networks ◽

Learning Models ◽

Healthcare Applications ◽

Ecg Signals ◽

Cardiac Abnormalities ◽

Adversarial Networks ◽

Data Points

Internet of Things (IoT) based healthcare applications have grown exponentially over the past decade. With the increasing number of fatalities due to cardiovascular diseases (CVD), it is the need of the hour to detect any signs of cardiac abnormalities as early as possible. This calls for automation on the detection and classification of said cardiac abnormalities by physicians. The problem here is that, there is not enough data to train Deep Learning models to classify ECG signals accurately because of sensitive nature of data and the rarity of certain cases involved in CVDs. In this paper, we propose a framework which involves Generative Adversarial Networks (GAN) to create synthetic training data for the classes with less data points to improve the performance of Deep Learning models trained with the dataset. With data being input from sensors via cloud and this model to classify the ECG signals, we expect the framework to be functional, accurate and efficient.

Download Full-text

GGADN: Guided Generative Adversarial Dehazing Network

10.21203/rs.3.rs-386958/v1 ◽

2021 ◽

Author(s):

Zhang Jian ◽

Wanjuan Song

Keyword(s):

Deep Learning ◽

Loss Function ◽

Minimax Problem ◽

Feature Model ◽

Generative Adversarial Networks ◽

Image Dehazing ◽

Learning Methods ◽

Adversarial Networks ◽

Network Training

Abstract Image dehazing has always been a challenging topic in image processing. The development of deep learning methods, especially the Generative Adversarial Networks(GAN), provides a new way for image dehazing. In recent years, many deep learning methods based on GAN have been applied to image dehazing. However, GAN has two problems in image dehazing. Firstly, For haze image, haze not only reduces the quality of the image, but also blurs the details of the image. For Gan network, it is difficult for the generator to restore the details of the whole image while removing the haze. Secondly, GAN model is defined as a minimax problem, which weakens the loss function. It is difficult to distinguish whether GAN is making progress in the training process. Therefore, we propose a Guided Generative Adversarial Dehazing Network(GGADN). Different from other generation adversarial networks, GGADN adds a guided module on the generator. The guided module verifies the network of each layer of the generator. At the same time, the details of the map generated by each layer are strengthened. Network training is based on the pre-trained VGG feature model and L1-regularized gradient prior which is developed by new loss function parameters. From the dehazing results of synthetic images and real images, proposed method is better than the state-of-the-art dehazing methods.

Download Full-text

GENERATING SYNTHETIC TRAINING DATA FOR OBJECT DETECTION USING MULTI-TASK GENERATIVE ADVERSARIAL NETWORKS

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-v-2-2020-443-2020 ◽

2020 ◽

Vol V-2-2020 ◽

pp. 443-449

Author(s):

Y. Lin ◽

K. Suzuki ◽

H. Takeda ◽

K. Nakamura

Keyword(s):

Neural Network ◽

Deep Learning ◽

Rapid Development ◽

Real Data ◽

Training Data ◽

Generative Adversarial Networks ◽

Great Success ◽

Traffic Signs ◽

Adversarial Networks ◽

The Neural Network

Abstract. Nowadays, digitizing roadside objects, for instance traffic signs, is a necessary step for generating High Definition Maps (HD Map) which remains as an open challenge. Rapid development of deep learning technology using Convolutional Neural Networks (CNN) has achieved great success in computer vision field in recent years. However, performance of most deep learning algorithms highly depends on the quality of training data. Collecting the desired training dataset is a difficult task, especially for roadside objects due to their imbalanced numbers along roadside. Although, training the neural network using synthetic data have been proposed. The distribution gap between synthetic and real data still exists and could aggravate the performance. We propose to transfer the style between synthetic and real data using Multi-Task Generative Adversarial Networks (SYN-MTGAN) before training the neural network which conducts the detection of roadside objects. Experiments focusing on traffic signs show that our proposed method can reach mAP of 0.77 and is able to improve detection performance for objects whose training samples are difficult to collect.

Download Full-text

Conditional Generative Adversarial Networks for Commonsense Machine Comprehension

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence ◽

10.24963/ijcai.2017/576 ◽

2017 ◽

Cited By ~ 6

Author(s):

Bingning Wang ◽

Kang Liu ◽

Jun Zhao

Keyword(s):

Reading Comprehension ◽

Deep Learning ◽

State Of The Art ◽

Training Data ◽

Generative Adversarial Networks ◽

Feature Engineering ◽

Language Understanding ◽

Story Reading ◽

Testing Period ◽

Adversarial Networks

Recently proposed Story Cloze Test [Mostafazadeh et al., 2016] is a commonsense machine comprehension application to deal with natural language understanding problem. This dataset contains a lot of story tests which require commonsense inference ability. Unfortunately, the training data is almost unsupervised where each context document followed with only one positive sentence that can be inferred from the context. However, in the testing period, we must make inference from two candidate sentences. To tackle this problem, we employ the generative adversarial networks (GANs) to generate fake sentence. We proposed a Conditional GANs in which the generator is conditioned by the context. Our experiments show the advantage of the CGANs in discriminating sentence and achieve state-of-the-art results in commonsense story reading comprehension task compared with previous feature engineering and deep learning methods.

Download Full-text

GANDaLF: GAN for Data-Limited Fingerprinting

Proceedings on Privacy Enhancing Technologies ◽

10.2478/popets-2021-0029 ◽

2021 ◽

Vol 2021 (2) ◽

pp. 305-322

Author(s):

Se Eun Oh ◽

Nate Mathews ◽

Mohammad Saidur Rahman ◽

Matthew Wright ◽

Nicholas Hopper

Keyword(s):

Deep Learning ◽

Training Data ◽

Generative Adversarial Networks ◽

Large Set ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Adversarial Networks ◽

Closed World ◽

Training Samples ◽

A Site

Abstract We introduce Generative Adversarial Networks for Data-Limited Fingerprinting (GANDaLF), a new deep-learning-based technique to perform Website Fingerprinting (WF) on Tor traffic. In contrast to most earlier work on deep-learning for WF, GANDaLF is intended to work with few training samples, and achieves this goal through the use of a Generative Adversarial Network to generate a large set of “fake” data that helps to train a deep neural network in distinguishing between classes of actual training data. We evaluate GANDaLF in low-data scenarios including as few as 10 training instances per site, and in multiple settings, including fingerprinting of website index pages and fingerprinting of non-index pages within a site. GANDaLF achieves closed-world accuracy of 87% with just 20 instances per site (and 100 sites) in standard WF settings. In particular, GANDaLF can outperform Var-CNN and Triplet Fingerprinting (TF) across all settings in subpage fingerprinting. For example, GANDaLF outperforms TF by a 29% margin and Var-CNN by 38% for training sets using 20 instances per site.

Download Full-text

Discriminative Sample Generation for Deep Imbalanced Learning

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence ◽

10.24963/ijcai.2019/334 ◽

2019 ◽

Cited By ~ 1

Author(s):

Ting Guo ◽

Xingquan Zhu ◽

Yang Wang ◽

Fang Chen

Keyword(s):

Deep Learning ◽

Class Imbalance ◽

Imbalanced Data ◽

Feature Learning ◽

Training Data ◽

Generative Adversarial Networks ◽

Component Mixture ◽

Minority Class ◽

Latent Features ◽

Class Labels

In this paper, we propose a discriminative variational autoencoder (DVAE) to assist deep learning from data with imbalanced class distributions. DVAE is designed to alleviate the class imbalance by explicitly learning class boundaries between training samples, and uses learned class boundaries to guide the feature learning and sample generation. To learn class boundaries, DVAE learns a latent two-component mixture distributor, conditioned by the class labels, so the latent features can help differentiate minority class vs. majority class samples. In order to balance the training data for deep learning to emphasize on the minority class, we combine DVAE and generative adversarial networks (GAN) to form a unified model, DVAAN, which generates synthetic instances close to the class boundaries as training data to learn latent features and update the model. Experiments and comparisons confirm that DVAAN significantly alleviates the class imbalance and delivers accurate models for deep learning from imbalanced data.

Download Full-text

Improved Semantic Segmentation of Water Bodies and Land in SAR Images Using Generative Adversarial Networks

International Journal of Semantic Computing ◽

10.1142/s1793351x20400036 ◽

2020 ◽

Vol 14 (01) ◽

pp. 55-69 ◽

Cited By ~ 3

Author(s):

M. M. Manohara Pai ◽

Vaibhav Mehrotra ◽

Ujjwal Verma ◽

Radhika M. Pai

Keyword(s):

Deep Learning ◽

Semantic Segmentation ◽

Ease Of Use ◽

Training Data ◽

Superior Performance ◽

Easy Access ◽

Generative Adversarial Networks ◽

Data Repository ◽

Sar Images ◽

Adversarial Networks

The availability of computationally efficient and powerful Deep Learning frameworks and high-resolution satellite imagery has created new approach for developing complex applications in the field of remote sensing. The easy access to abundant image data repository made available by different satellites of space agencies such as Copernicus, Landsat, etc. has opened various avenues of research in monitoring the world’s oceans, land, rivers, etc. The challenging research problem in this direction is the accurate identification and subsequent segmentation of surface water in images in the microwave spectrum. In the recent years, deep learning methods for semantic segmentation are the preferred choice given its high accuracy and ease of use. One major bottleneck in semantic segmentation pipelines is the manual annotation of data. This paper proposes Generative Adversarial Networks (GANs) on the training data (images and their corresponding labels) to create an enhanced dataset on which the networks can be trained, therefore, reducing human effort of manual labeling. Further, the research also proposes the use of deep-learning approaches such as U-Net and FCN-8 to perform an efficient segmentation of auto annotated, enhanced data of water body and land. The experimental results show that the U-Net model without GAN achieves superior performance on SAR images with pixel accuracy of 0.98 and F1 score of 0.9923. However, when augmented with GANs, the results saw a rise in these metrics with PA of 0.99 and F1 score of 0.9954.

Download Full-text

GANTL: Towards Practical and Real-Time Topology Optimization with Conditional GANs and Transfer Learning

Journal of Mechanical Design ◽

10.1115/1.4052757 ◽

2021 ◽

pp. 1-32

Author(s):

Mohammad Mahdi Behzadi ◽

Horea T. Ilies

Keyword(s):

Deep Learning ◽

Topology Optimization ◽

Knowledge Transfer ◽

Boundary Conditions ◽

Transfer Learning ◽

Computational Cost ◽

Generative Adversarial Networks ◽

Learning Methods ◽

Adversarial Networks ◽

Difficult Time

Abstract Many machine learning methods have been recently developed to circumvent the high computational cost of the gradient-based topology optimization. These methods typically require extensive and costly datasets for training, have a difficult time generalizing to unseen boundary and loading conditions and to new domains, and do not take into consideration topological constraints of the predictions, which produces predictions with inconsistent topologies. We present a deep learning method based on generative adversarial networks for generative design exploration. The proposed method combines the generative power of conditional GANs with the knowledge transfer capabilities of transfer learning methods to predict optimal topologies for unseen boundary conditions. We also show that the knowledge transfer capabilities embedded in the design of the proposed algorithm significantly reduces the size of the training dataset compared to the traditional deep learning neural or adversarial networks. Moreover, we formulate a topological loss function based on the bottleneck distance obtained from the persistent diagram of the structures and demonstrate a significant improvement in the topological connectivity of the predicted structures. We use numerous examples to explore the efficiency and accuracy of the proposed approach for both seen and unseen boundary conditions in 2D.

Download Full-text

Generative Adversarial Networks for Data Augmentation in Structural Adhesive Inspection

Applied Sciences ◽

10.3390/app11073086 ◽

2021 ◽

Vol 11 (7) ◽

pp. 3086

Author(s):

Ricardo Silva Peres ◽

Miguel Azevedo ◽

Sara Oleiro Araújo ◽

Magno Guedes ◽

Fábio Miranda ◽

...

Keyword(s):

Deep Learning ◽

Data Augmentation ◽

Manufacturing Sector ◽

Data Availability ◽

Training Data ◽

Generative Adversarial Networks ◽

Learning Approaches ◽

Adversarial Networks ◽

Novel Approach ◽

Structural Adhesive

The technological advances brought forth by the Industry 4.0 paradigm have renewed the disruptive potential of artificial intelligence in the manufacturing sector, building the data-driven era on top of concepts such as Cyber–Physical Systems and the Internet of Things. However, data availability remains a major challenge for the success of these solutions, particularly concerning those based on deep learning approaches. Specifically in the quality inspection of structural adhesive applications, found commonly in the automotive domain, defect data with sufficient variety, volume and quality is generally costly, time-consuming and inefficient to obtain, jeopardizing the viability of such approaches due to data scarcity. To mitigate this, we propose a novel approach to generate synthetic training data for this application, leveraging recent breakthroughs in training generative adversarial networks with limited data to improve the performance of automated inspection methods based on deep learning, especially for imbalanced datasets. Preliminary results in a real automotive pilot cell show promise in this direction, with the approach being able to generate realistic adhesive bead images and consequently object detection models showing improved mean average precision at different thresholds when trained on the augmented dataset. For reproducibility purposes, the model weights, configurations and data encompassed in this study are made publicly available.

Download Full-text

Improved SinGAN Integrated with an Attentional Mechanism for Remote Sensing Image Classification

Remote Sensing ◽

10.3390/rs13091713 ◽

2021 ◽

Vol 13 (9) ◽

pp. 1713

Author(s):

Songwei Gu ◽

Rui Zhang ◽

Hongxia Luo ◽

Mengyao Li ◽

Huamei Feng ◽

...

Keyword(s):

Remote Sensing ◽

Real Life ◽

Attention Mechanism ◽

Training Data ◽

Generative Adversarial Networks ◽

Natural Image ◽

Remote Sensing Images ◽

Training Time ◽

Adversarial Networks ◽

Remote Sensing Image Classification

Deep learning is an important research method in the remote sensing field. However, samples of remote sensing images are relatively few in real life, and those with markers are scarce. Many neural networks represented by Generative Adversarial Networks (GANs) can learn from real samples to generate pseudosamples, rather than traditional methods that often require more time and man-power to obtain samples. However, the generated pseudosamples often have poor realism and cannot be reliably used as the basis for various analyses and applications in the field of remote sensing. To address the abovementioned problems, a pseudolabeled sample generation method is proposed in this work and applied to scene classification of remote sensing images. The improved unconditional generative model that can be learned from a single natural image (Improved SinGAN) with an attention mechanism can effectively generate enough pseudolabeled samples from a single remote sensing scene image sample. Pseudosamples generated by the improved SinGAN model have stronger realism and relatively less training time, and the extracted features are easily recognized in the classification network. The improved SinGAN can better identify sub-jects from images with complex ground scenes compared with the original network. This mechanism solves the problem of geographic errors of generated pseudosamples. This study incorporated the generated pseudosamples into training data for the classification experiment. The result showed that the SinGAN model with the integration of the attention mechanism can better guarantee feature extraction of the training data. Thus, the quality of the generated samples is improved and the classification accuracy and stability of the classification network are also enhanced.

Download Full-text