Reg-Gan: Semi-Supervised Learning Based on Generative Adversarial Networks for Regression

Rezagholiradeh, Mehdi; Haidar, Akmal

doi:10.1109/icassp.2018.8462534

Cited by 33 publications

(27 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the objective of distinguishing being real and fake examples is weighted strongly enough, the network may devote larger portions of the network to the real/fake classification task, thereby reducing its effectiveness in the regression prediction. We show in our experiments that our proposed method outperforms the DG-GAN, both in our own implementation and in that of Rezagholiradeh and Haidar (2018).…”

Section: Alternative Semi-supervised Regression Gan Methodsmentioning

confidence: 76%

“…Even a small percentage of the images, when randomly chosen, will contain the primary attributes of a large portion of the dataset. However, for comparison purposes, we have followed the experimental procedure used by Rezagholiradeh and Haidar (2018). We have additionally provided results for significantly lower numbers of labeled images.…”

Section: Driving Steering Angle Predictionmentioning

confidence: 99%

“…The results of our method in comparison to the methods presented by Rezagholiradeh and Haidar (2018) are shown in Table 2. In these experiments, we show that our SR-GAN method significantly outperforms the Reg-GAN method for any number of labeled examples.…”

Section: Driving Steering Angle Predictionmentioning

confidence: 99%

See 2 more Smart Citations

Generalizing semi-supervised generative adversarial networks to regression using feature contrasting

Olmschenk

Zhu

Tang

2019

Computer Vision and Image Understanding

View full text Add to dashboard Cite

In this work, we generalize semi-supervised generative adversarial networks (GANs) from classification problems to regression problems. In the last few years, the importance of improving the training of neural networks using semi-supervised training has been demonstrated for classification problems. We present a novel loss function, called feature contrasting, resulting in a discriminator which can distinguish between fake and real data based on feature statistics. This method avoids potential biases and limitations of alternative approaches. The generalization of semi-supervised GANs to the regime of regression problems of opens their use to countless applications as well as providing an avenue for a deeper understanding of how GANs function. We first demonstrate the capabilities of semi-supervised regression GANs on a toy dataset which allows for a detailed understanding of how they operate in various circumstances. This toy dataset is used to provide a theoretical basis of the semi-supervised regression GAN. We then apply the semi-supervised regression GANs to a number of real-world computer vision applications: age estimation, driving steering angle prediction, and crowd counting from single images. We perform extensive tests of what accuracy can be achieved with significantly reduced annotated data. Through the combination of the theoretical example and real-world scenarios, we demonstrate how semi-supervised GANs can be generalized to regression problems.

show abstract

Section: Alternative Semi-supervised Regression Gan Methodsmentioning

confidence: 76%

Section: Driving Steering Angle Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Generalizing semi-supervised generative adversarial networks to regression using feature contrasting

Olmschenk

Zhu

Tang

2019

Computer Vision and Image Understanding

View full text Add to dashboard Cite

show abstract

“…They used daily stock data to train the model and succeeded in achieving improved forecasting accuracy. Rezagholiradeh et al [29] utilized GAN to resolve a regression problem. They modified the GAN structure such that their GAN model can simultaneously generate training data and perform forecasting.…”

Section: A Related Workmentioning

confidence: 99%

Conditional Tabular GAN-Based Two-Stage Data Generation Scheme for Short-Term Load Forecasting

et al. 2020

View full text Add to dashboard Cite

“…While the SSL techniques mentioned above show good performance, they have been mostly verified only for datasets with relatively low classification difficulty, such as CIFAR10 [22] and SVHN [23]. Furthermore, there are not many cases where SSL has been applied to a specific application [7,21,[37][38][39]. For example, [7] applied SSL to 3D CNN for FER.…”

Section: ) Applications Of Sslmentioning

confidence: 99%

Semi-Supervised Learning for Continuous Emotion Recognition Based on Metric Learning

Choi

Song

2020

IEEE Access

View full text Add to dashboard Cite

Emotion recognition is important for the interaction between human and artificial intelligence. Recently, the performance of facial image-based emotion recognition has been improved with deep learning's power. Nonetheless, huge data and label information for training are burdensome. In particular, annotating emotion labels in the continuous domain is very costly. Thus, we propose a novel semi-supervised learning that can not only reduce the annotation cost, but also improve emotion recognition performance by training with additional unlabeled data. The proposed method employs deep metric learning to improve feature embedding performance. Also, pseudo labels of unlabeled data are produced by analyzing inter-data distance in the feature space. Since pseudo labeling makes unlabeled data trainable, it increases overall performance. The experimental results show that the proposed method provides outstanding performance in the well-known MAHNOB-HCI dataset and the INHA dataset produced by our research team.

show abstract

Reg-Gan: Semi-Supervised Learning Based on Generative Adversarial Networks for Regression

Cited by 33 publications

References 3 publications

Generalizing semi-supervised generative adversarial networks to regression using feature contrasting

Generalizing semi-supervised generative adversarial networks to regression using feature contrasting

Conditional Tabular GAN-Based Two-Stage Data Generation Scheme for Short-Term Load Forecasting

Semi-Supervised Learning for Continuous Emotion Recognition Based on Metric Learning

Contact Info

Product

Resources

About