Semi-supervised classification model based on Triple Generative Adversarial Networks for Weather Forecasting

Recently, many studies have reported on image synthesis based on Generative Adversarial Networks (GAN). However, the use of GAN does not provide much attention on the signal classification problem. In the context of using wireless signals to classify illegal Unmanned Aerial Vehicles (UAVs), this paper explores the feasibility of using GAN to improve the training datasets and obtain a better classification model, thereby improving the accuracy of classification. First, we use the generative model of GAN to generate a large datasets, which does not need manual annotation. At the same time, the discriminative model of GAN is improved to classify the types of signals based on the loss function of the discriminative model. Finally, this model can be used to the outdoor environment and obtain a real-time illegal UAVs signal classification system. Our experiments confirmed that the improvements on the Auxiliary Classifier Generative Adversarial Networks (AC-GANs) by limited datasets achieve excellent results. The recognition rate can reach more than 95% in the indoor environment, and this method is also applicable in the outdoor environment. Moreover, based on the theory of Wasserstein GANs (WGAN) and AC-GANs, a more robust Auxiliary Classifier Wasserstein GANs (AC-WGANs) model is obtained, which is suitable for multi-class UAVs. Through the combination of AC-WGANs and Universal Software Radio Peripheral (USRP) B210 software defined radio (SDR) platform, a real-time UAVs signal classification system is also implemented.

Download Full-text

Parallel Wavegan: A Fast Waveform Generation Model Based on Generative Adversarial Networks with Multi-Resolution Spectrogram

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) ◽

10.1109/icassp40776.2020.9053795 ◽

2020 ◽

Cited By ~ 3

Author(s):

Ryuichi Yamamoto ◽

Eunwoo Song ◽

Jae-Min Kim

Keyword(s):

Generative Adversarial Networks ◽

Generation Model ◽

Model Based ◽

Adversarial Networks ◽

Waveform Generation

Download Full-text

CAGAN: Consistent Adversarial Training Enhanced GANs

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

10.24963/ijcai.2018/359 ◽

2018 ◽

Cited By ~ 1

Author(s):

Yao Ni ◽

Dandan Song ◽

Xi Zhang ◽

Hao Wu ◽

Lejian Liao

Keyword(s):

Neural Network ◽

Parameter Space ◽

Supervised Classification ◽

State Of The Art ◽

Generative Adversarial Networks ◽

Image Generation ◽

Real Samples ◽

Adversarial Networks ◽

Novel Approach ◽

Adversarial Training

Generative adversarial networks (GANs) have shown impressive results, however, the generator and the discriminator are optimized in finite parameter space which means their performance still need to be improved. In this paper, we propose a novel approach of adversarial training between one generator and an exponential number of critics which are sampled from the original discriminative neural network via dropout. As discrepancy between outputs of different sub-networks of a same sample can measure the consistency of these critics, we encourage the critics to be consistent to real samples and inconsistent to generated samples during training, while the generator is trained to generate consistent samples for different critics. Experimental results demonstrate that our method can obtain state-of-the-art Inception scores of 9.17 and 10.02 on supervised CIFAR-10 and unsupervised STL-10 image generation tasks, respectively, as well as achieve competitive semi-supervised classification results on several benchmarks. Importantly, we demonstrate that our method can maintain stability in training and alleviate mode collapse.

Download Full-text

Generative Adversarial Positive-Unlabelled Learning

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

10.24963/ijcai.2018/312 ◽

2018 ◽

Cited By ~ 6

Author(s):

Ming Hou ◽

Brahim Chaib-draa ◽

Chao Li ◽

Qibin Zhao

Keyword(s):

Neural Networks ◽

Theoretical Analysis ◽

Supervised Classification ◽

Deep Neural Networks ◽

Generative Adversarial Networks ◽

Discriminative Learning ◽

Decision Boundary ◽

Negative Data ◽

New Paradigm ◽

Adversarial Networks

In this work, we consider the task of classifying binary positive-unlabeled (PU) data. The existing discriminative learning based PU models attempt to seek an optimal reweighting strategy for U data, so that a decent decision boundary can be found. However, given limited P data, the conventional PU models tend to suffer from overfitting when adapted to very flexible deep neural networks. In contrast, we are the first to innovate a totally new paradigm to attack the binary PU task, from perspective of generative learning by leveraging the powerful generative adversarial networks (GAN). Our generative positive-unlabeled (GenPU) framework incorporates an array of discriminators and generators that are endowed with different roles in simultaneously producing positive and negative realistic samples. We provide theoretical analysis to justify that, at equilibrium, GenPU is capable of recovering both positive and negative data distributions. Moreover, we show GenPU is generalizable and closely related to the semi-supervised classification. Given rather limited P data, experiments on both synthetic and real-world dataset demonstrate the effectiveness of our proposed framework. With infinite realistic and diverse sample streams generated from GenPU, a very flexible classifier can then be trained using deep neural networks.

Download Full-text