All Questions
Tagged with neural-networks generative-models
97 questions
0
votes
0
answers
16
views
Why do VAEs work?
I am currently reading into Variational Autoencoders, and although I kind of understand the mathematical background described in the original paper (Auto-encoding Variational Bayes), I am struggling ...
1
vote
0
answers
65
views
Is it possible to explain regression or classification, interpolation and generation using a single model structure?
Neural network is established as an universal approximator of all machine learning models. Further, double descent phenomenon in a neural network propagates the journey of regression to interpolation ...
- 81
0
votes
0
answers
35
views
How to achieve Voice Conversion Using Voice Samples of a Specific Person using any voice as input?
I'm working on a project involving voice conversion, aiming to transform a voice to sound like a specific person speaking Darija (a Moroccan Arabic dialect). I have collected a set of voice samples ...
- 11
1
vote
1
answer
57
views
NeRF vs mesh for text-to-3d generation
There seem to be multiple aproaches to generating 3d objects from text prompt. What's confusing is that some of them are generating NeRFs (https://arxiv.org/pdf/2308.16512), other's are generating ...
- 196
4
votes
1
answer
1k
views
How to rewrite DreamBooth loss in terms of $\epsilon$-prediction?
I'm trying to make the loss used in DreamBooth paper explicit, writing it in terms of the noise, as it is commonly written in the original diffusion article [1], instead of the image reconstruction ...
- 505
2
votes
0
answers
174
views
Maximum Mean Discrepancy (MMD) implementation as a metric to measure GAN performance [closed]
I am trying to evaluate the performance of the GAN model, I have trained. I found that there exist two major choices FID (Fréchet inception distance) and MMD (Maximum Mean Discrepancy) for comparing ...
- 21
3
votes
2
answers
897
views
How is the variance for a diffusion kernel derived for a diffusion model?
So I'm watching this video tutorial from CVPR this year on diffusion models, and I am confused by the variance term in the distribution on the left on the video. I understand that in the forward ...
- 31
1
vote
1
answer
459
views
What kind of Neural Networks are required for Diffusion models?
It appears that regular feed-forward and convolutions are not enough to make diffusion models work (from some personal limited testing, they do not work at all). The typical infrastructure was a U-Net ...
- 768
1
vote
1
answer
2k
views
When is my Wasserstein GAN-GP overfitting?
I have a hard time interpreting the WGAN-GP losses attached. At which epoch is D and/or G overfitting? The quality improves a lot overtime, yet the generator loss at later epochs does not appear on ...
0
votes
1
answer
89
views
In factor VAE, do you freeze the discriminator weights during the back propagation step for the FVAE loss?
In factor vae, Disentangling by Factorising, there are two losses that are minimized. One is the VAE loss (eq. 2 in the paper) that includes (1) reconstruction loss, (2) KL divergence and (3) Total ...
- 33
2
votes
1
answer
1k
views
How to generate new data with a VAE?
I have built the following function which takes as input some data and runs a VAE on them:
...
- 531
3
votes
2
answers
2k
views
Can the dimension of the latent space in VAEs, be larger than the dimension of the data?
I am experimenting with VAEs. There, there is a parameter that you need pass when you create the NN, which is the dimension of the latent space.
In the typical ...
- 531
0
votes
1
answer
114
views
VAE active units
According to Burda et al (2015) number of active units is computed as:
$ Cov_x(E_{z \sim q_\phi(z|x)}) > \delta $
for some particular delta. In the paper it is set to 0.02 empirically. But this ...
2
votes
1
answer
557
views
How do I resize output of a generative network to reconcile with the size of the real data?
I am designing a Generative Adversarial Network (GAN) trained on an image dataset. It has two components: the generator and the discriminator. The generative network outputs an artificial image. The ...
- 218
0
votes
1
answer
102
views
Normalizing Flows Invertibility
I am currently reading up on RealNVP, which has the following transformations according Lilian Weng:
$$
\begin{aligned}
\mathbf{y}_{1:d} &= \mathbf{x}_{1:d} \\
\mathbf{y}_{d+1:D} &= \mathbf{x}...
- 19
2
votes
1
answer
967
views
Initialization of GAN discriminator
The question is pretty straightforward: how are GAN and WGAN discriminators typically initialized?
I couldn't find much info on this. E.x. for GANs, I imagine you would theoretically want the ...
- 143
2
votes
1
answer
480
views
Can PCA generate a new random image?
I read The Batch: GANs newsletter and Goodfellow said:
My colleague Bing Xu modeled face images from the Toronto Face Database, which were only 90 pixels square and grayscale. Because the faces were ...
- 121
22
votes
4
answers
8k
views
Likelihood-free inference - what does it mean?
Recently I have become aware of 'likelihood-free' methods being bandied about in literature. However I am not clear on what it means for an inference or optimization method to be likelihood-free.
In ...
- 967
3
votes
1
answer
444
views
Can someone explain CAM loss used in U-GAT-IT paper?
I have been reading a recent paper accepted at ICLR, U-GAT-IT, which seems to produce pleasing results in the image-to-image translation tasks. There are four kinds of loss used in this paper: ...
3
votes
3
answers
2k
views
In VAE, why use MSE loss between input x and decoded sample x' from latent distribution?
Variational Autoencoders (VAEs) are based on the concept of Variational Inference (VI) and use two Neural Networks similar to Vanilla Autoencoders (AEs) for function approximation. I understood the ...
- 61
0
votes
0
answers
244
views
What's GAN's Input-size Limitations?
I am interested in GAN for generating synthetic data. I am studying the input limitations for GAN starting from which GAN is no longer usable.
I have found many applications that use GANs for ...
- 101
1
vote
0
answers
389
views
Classical VAE not learning 2D gaussian mixture distribution using MSE loss
I've been exploring VAE for non-image data. I consider small to medium-sized continuous vector spaces and I want to learn the distribution of a dataset in that space.
As a warm up exercise, I tried ...
- 113
13
votes
3
answers
19k
views
Why use Binary Cross Entropy for Generator in Adversarial Networks
I'm trying to work with General Adversarial Networks and there's something I'm seeing everywhere but can't explain why...
the GANs are usually constructed from a Generator (which usually generates an ...
- 515
3
votes
1
answer
209
views
Why is using a mixture of logstic distributions makes sense in pixelcnn++?
I went trough the paper and code of the pixelcnn++ model.
From what I understand, they train the network in the following way for predicting the value of a single pixel:
the inputs are the pixel ...
- 273
3
votes
2
answers
7k
views
What is the stop criteria of generative adversarial nets?
I have used the GANs (Generative Adversarial Networks) with a binary cross-entropy loss function (in both generator and discriminator). Throughout the training step, the variation of generator loss ...
- 51
14
votes
3
answers
3k
views
Is the optimization of the Gaussian VAE well-posed?
In a Variational Autoencoder (VAE), given some data $x$ and latent variables $t$ with prior distribution $p(t) = \mathcal{N}(t \mid 0, I)$, the encoder aims to learn a distribution $q_{\phi}(t)$ that ...
- 707
3
votes
0
answers
464
views
If neural networks are deterministic once trained how are generative able to generate different responses to inputs?
Once training is finished and a neural network is in inference mode, its outputs will be deterministic, i.e. the weights have been fixed. How is it then that generative models are able to generate non-...
0
votes
0
answers
418
views
I need help understanding the meaning of the loss values of a WGAN with Gradient Penalty
I am currently working on training a Auxiliary Classifier Wasserstein GAN with Gradient Penalty. I based my implementation off of https://keras.io/examples/generative/wgan_gp/ (to which I added the ...
2
votes
0
answers
105
views
Restricted Boltzmann Machine: W matrix visualization results after training MNIST images and Pseudo-log-likelihood
I am implementing RBM from scratch using Tensorflow and after training my RBM on the MNIST dataset for 200 epochs using Persistent CD with two steps of contrastive divergence, I learn the weights W ...
- 21
13
votes
1
answer
730
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, ...
- 235
2
votes
1
answer
297
views
Issues with GAN and VAE models
I'm reading this amazing paper on Normalizing Flows https://arxiv.org/pdf/1908.09257.pdf but one sentence kind of bothers me:
GANs and VAEs have demonstrated impressive performance results on
...
22
votes
1
answer
6k
views
Why in Variational Auto Encoder (Gaussian variational family) we model $\log\sigma^2$ and not $\sigma^2$ (or $\sigma$) itself?
In theory the encoder in VAE (assuming that variational family is Gaussian) generates the $\mu$ and $\sigma$ (or $\sigma^2$). But, in practice, I have seen people assuming the output is $\log\sigma^2$....
- 1,502
16
votes
1
answer
18k
views
Variational Autoencoder − Dimension of the latent space
I've done some experiments to understand the influence of the dimension of the latent space in a VAE, and it seems that the higher the space, the harder it is to generate realistic images. I might ...
- 263
3
votes
1
answer
1k
views
How to find bits/dim of a gaussian output distribution?
I have images that are 64x64x3 and 64x64x1 8-bit. I transform those images down to [-1,1]. I now want to find the bits/dim for my VAE log probability. How do I find the bits/dim of the log likelihood? ...
- 65
2
votes
1
answer
3k
views
A meaning/significance of validation loss in a Generative Adversarial Neural Network? [closed]
On most of the tutorials on GANs that I came across the only monitored quantity is training loss.
1) Are there any general conclusions that could be derived from comparing training and validation ...
- 43
4
votes
2
answers
2k
views
Variational Autoencoder (VAE) latent features
I'm new to DL and I'm working on VAE for biomedical images. I need to extract relevant features from ct scan. So I created first an autoencoder and after a VAE. My doubt is that I don't know from ...
6
votes
3
answers
1k
views
What can possibly go wrong in a Generative Adversarial Network?
Lately, after reading about GANs, I started experimenting with the MNIST dataset, and the result we acceptable. Here are some details about the networks I used:
Discriminator: 784 inputs $\rightarrow$...
- 223
7
votes
2
answers
4k
views
Wasserstein Loss is very sensitive to model architecture
I am working on a class project where I compare the performance of GAN and WGAN. Since the only difference between GAN and WGAN is the Wasserstein loss, I chose one neural network model architecture ...
- 193
0
votes
1
answer
48
views
Is there such a thing as intra-sample modal collapse in GANs?
Mode collapse is a known issue in generative adversarial networks (GANs) whereby the generator only learns a subset of the real data distribution. In those cases, it only outputs variations of a small ...
- 169
2
votes
2
answers
414
views
What concept comes before VAE and after GMM?
Suppose I am designing a course on generative models and I have just finished discussing GMM. My goal is to teach VAE.
However, VAE's technicality is very high. Does there exist some model in between ...
- 734
1
vote
1
answer
2k
views
Understanding reparameterization trick and training process in variational autoencoders
I am trying to understand variational autoencoders, particularly the sampling component and the reparameterization trick. I understand that instead of using a fixed determinstic latent representation ...
- 1,030
8
votes
1
answer
1k
views
Clarification: Are Generative Adversarial Networks an alternative to MCMC sampling?
I have been reading the original Goodfellow, et. al. paper on Generative Adversarial Networks and the way that they can obtain estimates of the posterior distribution of a discriminative network or ...
- 1,582
0
votes
1
answer
2k
views
What is the Intuition behind the GAN Discriminator loss? How does Discriminator loss works?
I have just stated learning GAN and the loss used are different for same problems in same tutorial. Could someone please tell me intutively that which loss function is doing what?
For example, in the ...
- 244
5
votes
3
answers
361
views
Tutorial recommendations for understanding GANs
I understand that generative adversarial networks (GANs) can synthetically reconstruct the input using a generator and a discriminator in a zero-sum game. However, I feel that I do not fully ...
19
votes
2
answers
6k
views
Modern Use Cases of Restricted Boltzmann Machines (RBM's)?
Background: A lot of the modern research in the past ~4 years (post alexnet) seems to have moved away from using generative pretraining for neural networks to achieve state of the art classification ...
- 625
1
vote
0
answers
220
views
Synthetic data generation - GANs vs Simulator?
For synthetic data generation, does the GAN perform better than a simulator? If so, what are the limitations of the simulator?
If we consider Conditional GANs, we could generate data based on the ...
- 111
2
votes
1
answer
2k
views
On evaluating variational autoencoders with prior likelihood and reconstruction error
A common evaluation metric for variational autoencoders (VAEs) is estimating the marginal likelihood of some held-out data, i.e. $p(x)$. This is difficult and often one can only get a lower bound. It'...
- 2,284
1
vote
1
answer
50
views
Is the density of PixelCNN normalized?
PixelCNN++ constructs a model distribution $p(x)$ over images $x\in\mathbb{R}^{n\times n}$ as a product of conditional distributions over pixels
$$p(x)=p(x_1,...,x_{n^2})=\prod_{i=1}^{n^2} p(x_i| x_1,....
5
votes
1
answer
7k
views
How to interpret the following GAN training losses?
I am training a GAN using the following loss functions:
...
- 171
12
votes
2
answers
7k
views
What does log-likelihood mean in the context of generative models like GANs?
I understand the general notion of likelihood as "probability to generate the data given parameters" (like here). But people use (log-)likelihood as a measure of "goodness" of a generative model.
...
- 223