Questions tagged [generative-adversarial-networks]

For questions related to generative adversarial networks (GANs), introduced in the paper Generative Adversarial Nets (2014) by J. Goodfellow et al. A GAN is composed of a discriminative model (D) and a generative model (G). The discriminator D needs to distinguish between data generated by the generator G and data in the training set, while the generator G needs to generate data such that the discriminator D is not able to accomplish its task.

Filter by
Sorted by
Tagged with
2 votes
1 answer
61 views

What is being optimized with WGAN loss? Is the generator maximizing or minimizing the critic value?

I am kind of new to the field of GANs and decided to develop a WGAN. All of the information online seems to be kind of contradicting itself. The more I read, the more I become confused, so I'm hoping ...
user avatar
1 vote
0 answers
37 views

Is there a name for this model?

I have an image autoencoder model trained as follows: Step 1) train a GAN to obtain a generator capable of drawing from the data manifold by sampling a normal distribution in latent space Step 2) ...
user avatar
0 votes
1 answer
39 views

What is the meaning of $p_{\text {data }}(y)$ in the CycleGAN?

In the original CycleGAN paper, on the second page, there is a sentence that I didn't quite understand In theory, this objective can induce an output distribution over $\hat{y}$ that matches the ...
user avatar
0 votes
2 answers
63 views

What is the domain of the discriminator of a GAN?

I've read that the discriminator $D$ validates an image $D(x)$, where $x$ is either a real image or a fake one created by the generator, i.e. $ D(G(x))$. What does the function of the discriminator ...
user avatar
0 votes
1 answer
28 views

Why do we train the discriminators k times but train the generator only 1 time in a iteration in GAN?

In this paper https://arxiv.org/abs/1406.2661 , the codes for training a gan are : Why do we train the discriminator for $k$ steps while the generator only for $1$ step? Why not the other way around?
user avatar
  • 3
0 votes
0 answers
51 views

How to combine an image and a set of parameters as input to a GAN?

I have worked with a few GAN-like algorithms, but always with similar inputs and outputs. Being only a novice in deep learning, I often work by adapting an already existing Notebook, but today I have ...
user avatar
0 votes
0 answers
14 views

Why even-sized kernels are used in upscaling layers?

I have noticed that UNet and many GANs uses even-sized kernels in the upscale part of the model. I have read that at least in the GAN situation one of the reasons why we use even-sized kernels is that ...
user avatar
2 votes
1 answer
38 views

Why do we use the same parameters for the joint, marginal and conditional distributions in VAEs?

I've noticed in several resources on variational autoencoders (for example the wikipedia article), we use the same parameters theta for the prior, likelihood, posterior, etc distributions. For example ...
user avatar
  • 21
0 votes
0 answers
20 views

What are the benefits of using multiple convolutions, as opposed to one, before the pooling layer in a U-Net?

I have seen U-Nets that use a single convolution before the pooling operator and some that use two or more. My question is, what is better? Or what are the benefits of using more or less convolutions?
user avatar
0 votes
1 answer
22 views

Is the "Helvetica scenario" mentioned here related to Artificial Intelligence?

Consider the following sentence from the original GAN paper titled Generative Adversarial Nets in particular, $G$ must not be trained too much without updating $D$, in order to avoid "the ...
user avatar
  • 3,099
0 votes
0 answers
19 views

Why GAN is not as useful in tabular settings

This is inspired by a previous question Why are GAN models not heavily used for NLP? One answer says that the GAN is not as useful outside the image domain. Recently my friend and I are trying it out ...
user avatar
0 votes
1 answer
48 views

Training a GAN after after evaluation metric reaches minimum

I am training a StyleGAN-3 using one of the pre-trained models. At some point, roughly halfway through the 5000 kimg recommended for fine-tuning, the FID50K score starts oscillating around a minimum ...
user avatar
  • 3
1 vote
0 answers
38 views

Does all GAN's in literature need to satisfy the properties of objective function of initial GAN? [closed]

Consider the following value function of the initial GAN $V(D, G) = \mathbb{E}_{x \sim p_{data(x)}} [\log D(x)] + \mathbb{E}_{z \sim p_z(z)} [1- \log D(G(z))]$ The min-max game on the value function: $...
user avatar
  • 3,099
3 votes
2 answers
46 views

Why doesn't the inception score measure intra-class diversity

It's mentioned here that there is no measure of intra-class diversity with the inception score: If your generator generates only one image per classifier image class, repeating each image many times, ...
user avatar
0 votes
0 answers
24 views

Finer control over the distribution of output from a GAN

I have cast some data into an image (25 x 56) to work better with existing tools, and then used a CNN to train a GAN using the Wasserstein loss with gradient penalty to generate new samples of the ...
user avatar
  • 1
0 votes
0 answers
12 views

How is state size obtained/calculated?

I am looking at the Generator portion of the Pytorch tutorial for Generative Adversarial Networks and I am confused as to how the last 2 dimensions are obtained for what is ...
user avatar
1 vote
1 answer
87 views

What are the steps to derive the original GAN loss function from the generalized version?

I am trying to understand how the loss function from the original GAN paper $$\min_{G} \max_{D} V(D, G)=\mathbb{E}_{\boldsymbol{x} \sim p_{\text {data }}(\boldsymbol{x})}[\log D(\boldsymbol{x})]+\...
user avatar
  • 13
0 votes
0 answers
68 views

How exactly is modulation and demodulation layer in StyleGAN2 implemented?

So from the paper Analyzing and Improving the Image Quality of StyleGAN We know that the naive way to implement the stylegan2 Conv2DMod is to compute the Style vector which has the dimension of ...
user avatar
2 votes
1 answer
97 views

Does a colour consistency loss in neural networks (cycleGAN) make sense?

My neural network takes an image as an input and outputs another image. It's the generator of a cycleGAN. I would like to add (to the discriminator loss, the ...
user avatar
0 votes
1 answer
51 views

GANs inputs normalized and generator only outputs in [-1; 1]

I'm currently coding a GAN on the dataset MNIST. I'm using the following code to transform my data: ...
user avatar
0 votes
0 answers
39 views

Interpretation of the loss function of Wasserstein GAN: the lower the better?

After following this interesting collection of tutorials for GANs https://www.youtube.com/playlist?list=PLhhyoLH6IjfwIp8bZnzX8QR30TRcHO8Va I've been playing around experimenting Wasserstein GAN with ...
user avatar
3 votes
1 answer
466 views

Why are GAN models not heavily used for NLP?

I am wondering why there has not been more usage of GANs for NLP. I know there has been research on the subject (The Google Scholar page for the subject is here). Are there any specific reasons why ...
user avatar
2 votes
1 answer
85 views

Why do we use a linear interpolation of fake and real data to penalize the gradient of discriminator in WGAN-GP

I'm trying to better frame/summarize the formulations and motivations behind Wasserstein GAN with gradient penalty, based on my understanding. For the basic GAN we are trying to optimize the following ...
user avatar
0 votes
1 answer
76 views

Which approach can I use to generate forged signatures from real ones?

I am in internship period and I'm working on a signature verification problem. This process needs real and forged signatures. All I have are the real signatures (like 30 signatures per person), and I ...
user avatar
0 votes
1 answer
40 views

What is intended for 1x1 convolution for input images?

I'm reading this about Self-Attention GANs : https://sthalles.github.io/advanced_gans/ I'm trying to better frame what is intended about 1x1convolution for input ...
user avatar
1 vote
2 answers
39 views

How is the output of the Generator in a GAN corrected?

If the Generator in a GAN is taking a matrix of size WxH of noise to generate a WxH sized output image, and the Discriminator ...
user avatar
2 votes
1 answer
41 views

Is there a way to inject linear constrains during GAN training?

Given that I'm training a generative model, (say a generative adversarial network), and I know that my (real) inputs (let's say vectors $\textbf{x} \in \mathbb{R}^n$) satisfy linear constraints of the ...
user avatar
1 vote
1 answer
251 views

Are there some known neural networks that, given an input image, can generate a similar image, with the same topic?

Are there some known neural networks that, given an input image, can generate a similar image, with the same topic? Example: input = a photo of a cat on a green table, output = a generated photo of ...
user avatar
  • 49
0 votes
0 answers
49 views

Possible improvements to WGAN-GP output images

I am mapping rather complex data into what essentially amounts to a greyscale image to take better advantage of GANs for generative means. Here is an example of some real data: All real data is of ...
user avatar
  • 1
0 votes
0 answers
17 views

For each case, the input volume is scaled to [0.0, 1.0]

I taking up a project related to domain adaptation in medical imaging field. While I read a paper from 'Using Out-of-the-Box Frameworks for Contrastive Unpaired Image Translation for Vestibular ...
user avatar
0 votes
0 answers
13 views

Cognitive Sciences for No-Reference Perceptual Quality Assessment

I was reading on the theory of Generative Adversarial Networks, when I came upon the following article: How to Train your Generative Models by Ferenc Huszár. The following part left me with many ...
user avatar
0 votes
0 answers
61 views

GAN performance starts to get worse as training continues

I'm currently trying to train a GAN to recreate similar images from a dataset. The dataset is using the Eiffel Tower Pictures from Googles Quick Draw dataset. The images aren't very large (only 12x12 ...
user avatar
1 vote
0 answers
47 views

GANs: Why does iterative gradient descent sometimes optimise $\min_G \max_D V(D,G)$ and sometimes $\max_D \min_G V(D,G)$?

For the following minimax equation for generative adversarial networks (GANs), $$\min_G \max_D V(D,G) = \mathbb{E}_{\boldsymbol{x}\sim p_{data}(\boldsymbol{x})}[\log D(\boldsymbol{x})] + \mathbb{E}_{\...
user avatar
1 vote
1 answer
95 views

Why are logarithms used in GANs minimax equation?

The minimax equation for generative adversarial networks $$\min_G \max_D V(D,G) = \mathbb{E}_{\boldsymbol{x}\sim p_{data}(\boldsymbol{x})}[\log D(\boldsymbol{x})] + \mathbb{E}_{\boldsymbol{z}\sim p_{\...
user avatar
2 votes
1 answer
128 views

What does "Gau" in GauGAN stand for?

GauGAN is a neural network architecture from NVIDIA that can create realistic images from semantic maps (and nowadays also textual descriptions).
user avatar
0 votes
0 answers
34 views

What is the best way to generate handwritten text documents?

I am new to generative models. I was wondering if it would be better to generate an image of a handwritten text document as a whole (which I don't know how exactly is done), or first generate ...
user avatar
  • 235
0 votes
1 answer
52 views

Is a Conv2DTranspose the same as a full convolution?

I am currently creating a GAN model from scratch (following this tutorial: https://machinelearningmastery.com/how-to-develop-a-generative-adversarial-network-for-an-mnist-handwritten-digits-from-...
user avatar
0 votes
0 answers
28 views

Adversarial Autoencoder is not working and not learning properly

I am trying to get an Adversarial AutoEncoder going using keras Fit method on a keras.model class but for some reason it is not working. Keep in mind that I tried updating encoder and decoder at the ...
user avatar
  • 1
0 votes
0 answers
18 views

Improve generalization of phishing website detection with computer vision

I want to use computer vision to detect phishing websites. There has already been some study on this, which showed this is effective. Most phishing sites try to replicate well-known websites such as ...
user avatar
0 votes
2 answers
70 views

Validation set performance increasing even after seemingly overfit on training set

I am training a semi-supervised GAN network using data from multiple subjects. I separated the labeled and unlabeled data based on my subjects, so there is no leakage, while having much more unlabeled ...
user avatar
  • 31
0 votes
0 answers
131 views

Discrepencies between the TimeGan paper and the code?

I recently read the paper Time-Series Generative Neural Networks and found the results that they reported quite promising (https://proceedings.neurips.cc/paper/2019/file/...
user avatar
  • 41
1 vote
0 answers
20 views

How to understand the results of a generator that switches, for metric evaluation?

I am running a code on generative adversarial networks. The code is designed in such a way that it outputs a fake image after every 5 epochs. The total number of epochs is 800 in number. After the ...
user avatar
  • 3,099
0 votes
0 answers
28 views

How to assess the goodness of a text generation algorithm

Take a RNN network fed with Shakespeare and generating Shakespeare-like text. Once a model seems mathematically fine, as can be assessed by observing its loss and accuracy over training epochs, how ...
user avatar
  • 101
1 vote
0 answers
69 views

What is the confusion loss for adversarial learning?

What is the confusion loss used in domain adaptation (DA) for adversarial learning/GANs? See this paper. Two domains: $s$: source domain $t$: target domain Generator/Discriminator setting: $M_s:x_s\...
user avatar
  • 11
2 votes
1 answer
35 views

Is it impossible to evaluate the generator distribution directly?

The following excerpt is taken from 3. The Inception Score for Image Generation from the paper titled A Note on the Inception Score. Suppose we are trying to ...
user avatar
  • 3,099
0 votes
0 answers
41 views

What is meant by inverting the generator?

Generative Adversarial Networks, in general, consists of two multi layer perceptrons: generator and discriminator. Generator is used for generating samples that are as real as training samples and ...
user avatar
  • 3,099
0 votes
0 answers
29 views

How did authors ensure that critical points do exist in GAN?

Using an MLP as a generator introduces multiple critical points in parameter space. You can read this excerpt from the research paper titled Generative Adversarial Nets by Ian J. Goodfellow et al. In ...
user avatar
  • 3,099
1 vote
0 answers
44 views

When can we call a loss function "adaptive"?

A loss function is a measure of how bad our neural network is. We can decrease the loss by proper training. I came across the phrase "adaptive loss function" in several research papers. For ...
user avatar
  • 3,099
1 vote
0 answers
21 views

Is the following a typo or am I understanding wrongly regarding discriminator?

Consider the following paragraph from the section 3: Background of the research paper titled Generative Adversarial Text to Image Synthesis by Scott Reed et al. ...
user avatar
  • 3,099
0 votes
0 answers
14 views

Expression Transfer Deep Learning Problem

I have old video and I want to keep the person's face in the video but I want to transfer my facial expressions to that video. Is there any better alternative to first order motion model for that task ...
user avatar

1
2 3 4 5