DEGAS: differentiable efficient generator search

Doveh, Sivan; Giryes, Raja

doi:10.1007/s00521-021-06309-8

Cited by 15 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, in Table 7, we compare our method with the state-of-the-art unsupervised GANs. We should point out that these methods use different generators, as a result of network architecture search (Gong et al, 2019;Doveh & Giryes, 2019). Despite the non optimality of our generator architecture, our approach yields a very competitive FID.…”

Section: Methodsmentioning

confidence: 98%

A Unified Generative Adversarial Network Training via Self-Labeling and Self-Attention

Watanabe,

Favaro

2021

Preprint

View full text Add to dashboard Cite

We propose a novel GAN training scheme that can handle any level of labeling in a unified manner. Our scheme introduces a form of artificial labeling that can incorporate manually defined labels, when available, and induce an alignment between them. To define the artificial labels, we exploit the assumption that neural network generators can be trained more easily to map nearby latent vectors to data with semantic similarities, than across separate categories. We use generated data samples and their corresponding artificial conditioning labels to train a classifier. The classifier is then used to self-label real data. To boost the accuracy of the self-labeling, we also use the exponential moving average of the classifier. However, because the classifier might still make mistakes, especially at the beginning of the training, we also refine the labels through self-attention, by using the labeling of real data samples only when the classifier outputs a high classification probability score. We evaluate our approach on CIFAR-10, STL-10 and SVHN, and show that both self-labeling and self-attention consistently improve the quality of generated data. More surprisingly, we find that the proposed scheme can even outperform classconditional GANs.

show abstract

Section: Methodsmentioning

confidence: 98%

A Unified Generative Adversarial Network Training via Self-Labeling and Self-Attention

Watanabe,

Favaro

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Other approaches use reconstruction loss to optimize the generator without a discriminator [5,26]. Other approaches used architecture search to find the generator architecture [19,14,54,55].…”

Section: Related Workmentioning

confidence: 99%

GMM-Based Generative Adversarial Encoder Learning

Feigin¹,

Spitzer²,

Giryes³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

While GAN is a powerful model for generating images, its inability to infer a latent space directly limits its use in applications requiring an encoder. Our paper presents a simple architectural setup that combines the generative capabilities of GAN with an encoder. We accomplish this by combining the encoder with the discriminator using shared weights, then training them simultaneously using a new loss term. We model the output of the encoder latent space via a GMM, which leads to both good clustering using this latent space and improved image generation by the GAN. Our framework is generic and can be easily plugged into any GAN strategy. In particular, we demonstrate it both with Vanilla GAN and Wasserstein GAN, where in both it leads to an improvement in the generated images in terms of both the IS and FID scores. Moreover, we show that our encoder learns a meaningful representation as its clustering results are competitive with the current GAN-based state-of-the-art in clustering. 1

show abstract

“…The Progressive Growing GAN [21] was created based on [1,16] with the main idea of progressively adding new layers of higher resolution during training, which helps to create highly realistic images. [14,12,40] developed neural architecture search methods to find an optimal neural network architecture to train GAN for a particular task.…”

Section: Related Workmentioning

confidence: 99%

Adaptive Weighted Discriminator for Training Generative Adversarial Networks

Zadorozhnyy¹,

Cheng²,

Ye³

2020

Preprint

View full text Add to dashboard Cite

Generative adversarial network (GAN) has become one of the most important neural network models for classical unsupervised machine learning. A variety of discriminator loss functions have been developed to train GAN's discriminators and they all have a common structure: a sum of real and fake losses that only depends on the actual and generated data respectively. One challenge associated with an equally weighted sum of two losses is that the training may benefit one loss but harm the other, which we show causes instability and mode collapse. In this paper, we introduce a new family of discriminator loss functions that adopts a weighted sum of real and fake parts, which we call adaptive weighted loss functions, or aw-loss functions. Using the gradients of the real and fake parts of the loss, we can adaptively choose weights to train a discriminator in the direction that benefits the GAN's stability. Our method can be potentially applied to any discriminator model with a loss that is a sum of the real and fake parts. Experiments validated the effectiveness of our loss functions on an unconditional image generation task, improving the baseline results by a significant margin on CIFAR-10, STL-10, and CIFAR-100 datasets in Inception Scores and FID.

show abstract

DEGAS: differentiable efficient generator search

Cited by 15 publications

References 21 publications

A Unified Generative Adversarial Network Training via Self-Labeling and Self-Attention

A Unified Generative Adversarial Network Training via Self-Labeling and Self-Attention

GMM-Based Generative Adversarial Encoder Learning

Adaptive Weighted Discriminator for Training Generative Adversarial Networks

Contact Info

Product

Resources

About