Questions tagged [gan]

Generative Adversarial Networks (GANs) are neural networks that are trained in an adversarial manner to generate data mimicking some distribution.

Filter by
Sorted by
Tagged with
0
votes
0answers
10 views

Need to convert numpy arrayS to a single DatasetV1 Array [closed]

As of now, I have generated numpy arrays for X and y(data and labels) and am trying to feed them to thisisRon's cGAN code https://github.com/thisisiron/TF2-GAN/tree/master/cgan Part of his code(...
1
vote
0answers
26 views

Controlling details in images generated by Generative Adversarial Networks (GANs)

With conditional GANs it is possible to generate images of a certain class of objects. And moreover with current text-to-image methods it seems to be possible to control certain details of the ...
1
vote
0answers
24 views

Mathematical meaning of minimizing JS divergence about GAN [closed]

Optimization of the loss function of GAN is equivalent to minimizing Jensen Shannon divergence, and minimization of cross-entropy loss, which is often used in classification problems such as image ...
1
vote
0answers
41 views

How is vector arithmetic and interpolation possible in the latent space of GANs?

In the DCGAN paper (Alec Radford et al.), the authors were able to perform vector arithmetic for semantic analogies by averaging the latent vectors of generated images with the same class. They've ...
0
votes
0answers
23 views

Implementing the GAN loss function

I was reading the original Generative Adversarial Nets paper by Goodfellow et al. This is the minimax optimization problem for GANs: All the code I have seen, however, seem to have a different ...
1
vote
1answer
18 views

Does mislabeling due to adversarial noise in features count as adversarial machine learning?

According to the traditional definition, Adversarial machine learning is a technique employed in the field of machine learning which attempts to fool models through malicious input. However, I have ...
1
vote
0answers
7 views

How to standardize multiple time series for GAN

I have a set of 150 multivariate time series, each containing 10 variables measured at 50 time points. The goal is to generate artificial time series which are similar to the ones I have, so I trained ...
1
vote
1answer
29 views

GAN only generating a subset of image classes

Fairly new to training GANs, trying to train on MNIST using adversarially learned inference (based on this paper: https://arxiv.org/abs/1606.00704), and am running into the following problem: the ...
1
vote
1answer
58 views

Why is “weight clipping” needed for Wasserstein GANs?

I am reading the original paper on the Wasserstein GAN: https://arxiv.org/pdf/1701.07875.pdf and I came across this paragraph: I don't understand the statement: "$\mathcal{W}$ is compact implies ...
6
votes
2answers
87 views

What are the current methods to check for GAN overfitting?

In generative modeling, the goal is to find a way for a model to output samples of some distribution $p_X$ given a lot of samples $x_1, \ldots, x_n$. In particular, we want sampling from our model $G$ ...
0
votes
0answers
23 views

Probability distribution in GAN

Here is the general eauation of GAN. I have a problem regarding probability distribution of x and z. $$\min_G \max_D V(D,G) = \mathbb{E}_{x\sim p_{data}(x)}[\log D(x)]+ \mathbb{E}_{z\sim p_{z}(z)}[\...
4
votes
3answers
328 views

What methods can be used for distribution generation other than GANs?

Generative Adversarial Networks (GANs) can be used for creating distributions of data points, that follow source data set distributions (e.g. images, sound, text, etc). Are there any other methods or ...
0
votes
0answers
8 views

In GANs, why converging the discriminator to probability 0.5 does not make the generator to change its good learned weights?

It is said that the discriminator part of GANs converges to probability 0.5, meaning that when it gets an instance generated by generator, it says that it is a fake instance by probability equal to 0....
0
votes
0answers
14 views

Does discriminator converge to probability 0.5 in GAN just for the output of generator?

We know that in GANs, discriminator converges to 0.5. I want to become sure that the following statement is true: Discriminator converges to 0.5 for the examples generated by generator and for the ...
0
votes
1answer
24 views

Is it possible to use Generative Adversarial Networks (GANs) for text classification?

I am working on the classification of fake and real news. I did use a CNN Model for this problem and got satisfactory results. But, I was just wondering if it's at all possible to use any type of GAN ...
2
votes
2answers
25 views

How to extract crucial features to create an image

Imagine, you have a dataset containing pictures of (example only, just to explain the task) cats and dogs. The data set is labeled, so we can train using supervised learning algorithms. My goal is ...
-1
votes
2answers
68 views

GANs: What does the pdf of the sample data p(x) mean? [closed]

In the context of GANs, the concept of a probability distribution comes up as the generator tries to emulate the "distribution" of the data: $p_{data}(x)$. For me, the use of "distribution" here ...
0
votes
0answers
27 views

Is GAN effective enough to replace data augmentation and manual annotation?

We all know that GAN can be used to augment and expand our dataset Can a GAN be used for data augmentation?. But my question is, is it effective and fast enough? For example I have done experiment ...
0
votes
0answers
13 views

Is it possible to train a GAN discriminator based on a pre-trained GAN generator?

If you have an already-trained GAN generator but do not have the corresponding GAN discriminator, is it possible to train a new functionally equivalent discriminator from scratch without having to re-...
0
votes
0answers
38 views

Training a GAN and loss curves

I am training a GAN with the discriminator and the generator being trained alternately for 2 epochs each. As expected I end up with oscillating loss curves, but they aren't damping as the training ...
0
votes
0answers
18 views

How to draw CycleGANs discriminative distribution?

A few weeks ago I asked this question which was about how to understand GANs discriminative distribution. Now I am trying to draw a Figure like the one below, but for CycleGAN instead. In CycleGAN ...
2
votes
1answer
38 views

Can the GAN objective function be written as related to a log-likelihood of some “classical” statistical model?

Can the objective function that GAN (Generative Adversarial Network) models optimize be written as a lower bound of the log-likelihood of some "classical" statistical model? I am reading through the ...
0
votes
0answers
8 views

Generative Adversarial Networks evaluation methods for one channel

So I'm currently studying GANs with a focus on CycleGAN. I have trained my network on simulated images and real images. I did not train them as pairs but I have pairs available. The idea is now to ...
0
votes
1answer
57 views

Why noise with normal distribution is used as input to GAN?

Why noise with normal distribution is used as input to GAN? What will happen if we will use uniform noise or just random binary vector?
10
votes
1answer
489 views

How to understand Generative Adversarial Networks Discriminative distribution?

So I am currently studying Generative Adversarial Network and I read the paper by Goodfellow a few times now Generative Adversarial Nets and a few other papers in this field (DCGAN, CycleGAN, pix2pix, ...
1
vote
1answer
86 views

What does Goodfellow mean by “generator conditional variance”?

In Goodfellow's Generative Adversarial Nets, it is mentioned that Our work backpropagates derivatives through generative processes by using the observation that $$\lim_{\sigma \rightarrow 0} \...
0
votes
0answers
33 views

In context of neural network training for regression task, Should the continuous label be of uniform distribution?

Currently, I am using Deep Neural Network (DNN) for the regression task. The training data is time series and contains 7 input features and 836 training instances (samples). Here the label is ...
0
votes
0answers
69 views

Can GANs be used for timeseries data augmentation? (2019)

Timeseries, in particular signal timeseries, are distinct in many respects - so GANs working on images may not work for timeseries. Since other questions asking on data augmentation, GANs have ...
0
votes
0answers
10 views

discriminator function in dual Generative adversarial network

I am reading the following article: https://arxiv.org/pdf/1803.00385.pdf On page 3 they explain how the losses for two discriminators (D1 and D2) is determined. D1 separates true upright image from ...
0
votes
0answers
22 views

Combining a neural network classifier with genetic algorithm to do the task likes what generative adversarial network does?

Recently I read some materials about GANs(Generative adversarial network). I know that GANs uses a discriminator network to distinguish the expected output(let's say that it's an image),and a ...
0
votes
0answers
14 views

How to generate a path given a set of input features?

I have set of features for each element of my training set: features = [time, status, speed, on/off, ...]. Such features are just a single value for each element ...
0
votes
0answers
31 views

Can data generated by generative models be used for training unsupervised learning models? [duplicate]

I'm working on a signal denoising problem. Because of not having enough data for training I'm considering using one of the generative models like VAE or GAN to generate data similar to real data for ...
2
votes
1answer
33 views

Mode-collapse problem in GAN: How to grasp the full entropy of the distribution we want to model in GAN

In pix2pix GAN paper( https://arxiv.org/abs/1611.07004), authors found that the noise vector and the dropout are not efficient in grasping the full entropy of the data distribution we want to model. ...
0
votes
0answers
20 views

How to measure the quality of fakes produced by a digital twin?

I have asked this on stackoverflow, but it was considered too broad, perhaps because there is no specific code involved. I hope this question is more appropriate for crossvalidated. I am developing a ...
1
vote
1answer
27 views

Is it possible to obtain learn specific information about a distribution from a GAN?

If I feed a GAN some images of Gaussian noise with some $\sigma$ and it is successful at generating similar images, is there some way to recover $\sigma$ from the gan or is the generator purely a ...
0
votes
0answers
12 views

Implementing a specific GAN loss

In this paper in section 3.4 they define a GAN loss as (I simplify to the important part): $$\mathcal L_{GAN} = \max_D \min_G E_{x_{1} \sim P_1}[ \Vert D(G(x_{1})) \Vert] + \max_D E_{x_2 \sim P_2}[\...
1
vote
2answers
30 views

Working mechanism of discriminator in text to image synthesis GAN

I have the following architecture of discriminator in text to image synthesis where the image is convolved to lower dimension and concatenated with the text . My question is what is the use of ...
2
votes
1answer
132 views

How does one generate (smooth) varying size output signals with Machine Learning?

I am interested in knowing about generative methods that generate signals (e.g. images) of varying sizes. But the size generation being sort of "smooth/continuous". So for example, generating images ...
1
vote
0answers
98 views

calculate Inception Score and FID in GAN [closed]

Do we calculate Inception Score and Fréchet Inception Distance (FID) on the images generated by generator in parallel during training ? Or do we save the images generated by the generator and later ...
0
votes
0answers
12 views

What is this phenomenon? is this the so called “Model Collapse” phenomenon in a Gan or something completely different?

This is a follow up question which I asked here. I tried to see if my GAN memorized something or not. I'm 100% sure that in my earlier experiments I noticed couple of images that were identical to ...
1
vote
0answers
63 views

How does the number of epochs affecting GANs training?

In CNN training, increasing the number epochs would lead to overfitting. However, to train a GAN, would a too large number of training epochs matter? Indeed, I also do not understand what does it mean ...
2
votes
0answers
18 views

Generating distributions with very small values with generative adverserial networks

I am using a GAN to generate a vector distribution. However, most of the values of this distribution are very close to zero, i.e $1 \times 10^{-7}$. The distribution also have minus values. With ...
2
votes
0answers
249 views

GANs for non image data

I'm looking to narrow down the subject for my bachelor thesis: I am currently working on a project, that only offers a small dataset and there will be no more data incoming for now. What I'm trying to ...
1
vote
1answer
203 views

Increasing image size in pytorch celebrity generating GAN? [closed]

complete newbie here, bear with me. I'm making my way through this tutorial: https://pytorch.org/tutorials/beginner/dcgan_faces_tutorial.html Upon attempting to make a simple change to the image ...
3
votes
0answers
28 views

Distribution $f$ that minimizes $JSD(f||q) + JSD(f||p)$

What can we say about the distribution $f^*$ that is the solution to the following optimization problem: $$\min_f JSD(f||p)+JSD(f||q) ,$$ where $p,q$ are given distributions over some set, and $JSD$ ...
0
votes
1answer
40 views

Global Minimum solution for the Generator step in the original GAN formulation

I am going through the proof in Goodfellow's original GAN paper. Specifically, I'm at the stage where he did the discriminator step (solve a maximization problem) and now he is doing the generator ...
0
votes
0answers
24 views

Intuition for manifold-diffusion training

Manifold-diffusion training is introduced in the following paper: https://arxiv.org/pdf/1612.02136.pdf The basic idea is that when training a GAN, we try to accomplish two things, which the paper ...
11
votes
1answer
652 views

Notation: What does the tilde below of the expectation mean? [duplicate]

I am reading about variational auto encoders, and there is the below loss function: $$l_i(\Theta,\phi) = - {\mathbb{E}}_{z\sim q} \left[\log p_\phi(x_i|z)\right] + KL(q_{\phi}(z_i|x)||p(z))$$ What ...
1
vote
0answers
42 views

When to stop training a Auixilary Classifier GAN?

When should I stop training an Auxiliary Classifier GAN? when the discriminator loss converges or when the auxiliary classification accuracy converges or when the Real vs Fake classifier converges to ...
2
votes
0answers
23 views

How do I check my GAN implementation is correct?

I wrote a GAN implementation and I trained that to produce some sample images after training on a dataset. The images looked visually fine. Now I want to test my implementation on the CI and make ...