# Questions tagged [vae]

The tag has no usage guidance.

39 questions
Filter by
Sorted by
Tagged with
11 views

### Which is better KL- Divergence or Bhattacharya(Hellinger) Distance

I'm beginner in probability and statistics. I came across the concept of comparing two probability distributions. KL-Divergence and Bhattacharya(Hellinger) Distance are used to compare two probability ...
10 views

### How do I prevent infinite variances/standard deviations in my variational autoencoder?

I am working on a project with a variational autoencoder (VAE). The problem I have is that the encoder part of VAE is producing large log variances, which leads to even larger standard deviations, ...
39 views

### Latent variable graph in Variational Autoencoder

I followed this Keras documentation guide about Auto Encoders. At the end of the documentation there is the graph of the latent variable z: But I can not understand and how to interpret the plot, ...
4 views

### VAE will always results in somewhat different latent vectors for same input?

Hey I was wondering if my intuition is correct that for the same input in a VAE we will get a slightly different vector every time we feed it through the network, due to the random sampling operation?
14 views

### Fitting input data into Gaussian distribution

I'm currently reading papers on Variational Autoencoders (VAE). According to this article (http://proceedings.mlr.press/v95/guo18a/guo18a.pdf): By fitting the input data sample x(i) into the Gaussian ...
16 views

### Should we sample z in VAE encoder during inference, when used in RL pipeline like World Models?

My question has been motivated by reading World Models by Ha and Schmidhuber. In shortcut, they introduce a RL framework where the current state (an image) is encoded via VAE into a latent vector $z$, ...
11 views

### Issue when training VAE

I am trying to train a Variational Autoencoder (VAE) to learn a curve, $f_1(x)=x^2+\omega$. It can learn it to a great accuracy, but the problem is that once is trained, for any input to the encoder, ...
55 views

### Controlling the sampling from Variational AutoEncoder (VAE)

Suppose a Variational Autoencoder (VAE) is trained with mnist data. To sample, one draws from normal distribution. My question is: suppose I am interested in generating only 1s and no other digits. ...
32 views

### Help needed in interpreting the loss, val_loss vs epoch plots for an autoencoder training?

I am training a variational autoencoder and I am getting a loss-plot as follows: Rigt after epoch 224, val-loss overtakes train-loss and sort of getting bigger but at an extremely slow pace as you ...
6 views

### Can you approximate the empirical probability of a data-point from a trained Variational Autoencoder?

I'm interested in if it's possible to use a trained VAE that when you pass a specific data-point through it, the elbo can help approximate to the data-points empirical probability. If this is ...
6 views

### Training with different datasets for the same better VAE model yields poor results

The VAE model I used here https://github.com/keras-team/keras-io/blob/master/examples/generative/vae.py. It can produce very well results for the minist and fashion minist dataset. But when I use my ...
446 views

### 1D CNN Variational Autoencoder Conv1D Size

I am trying to create a 1D variational autoencoder to take in a 931x1 vector as input, but I have been having trouble with two things: Getting the output size of 931, since maxpooling and upsampling ...
13 views

### Variational autoencoders - encoder-decoder neural nets relationship to maximizing evidence

I am new to VAE and I do not know why if the encoder and decoder are learned by a neural networks then how is maximizing the ELBO (evidence lower bound) or maximizing evidence in general relevant to ...
30 views

### VAE KL-divergence with non-standard mean

I know I can make a VAE do generation with a mean of 0 and std-dev of 1. I tested it with the following loss function: ...
13 views

### Reversing data through a tensorflow feature_column.embedding_column

I am building an variational autoencoder in Tensorflow, and one of my columns has object data. The data is too sparse (on the order of 2^16 possible values) to use one-hot encoding, it's not ordinal ...
95 views

### VAE: first hidden layer of encoder having more neurons than the input layer

Many blog posts explain the architecture of variational autoencoders (VAEs) with symmetrical encoder-decoder architecture where the number of neurons in the encoder's hidden layers decreases, starting ...
14 views

### Single real number evaluation metric for VAE for a regression problem

I've set up a VAE for a regression problem. ...
12 views

### Weird cutdown in VAE mean graph

I'm experimenting with VAE with Tensorflow, and just following tutorial setups to have an approximate comparison. Well, they aren't similar at all. Tutorial mean graph: My mean graph: As you can see ...
34 views

### What are the Most Dissimilar MNIST Digits?

Using whatever definition of dissimilarity over sets that you'd like, what are the most dissimilar two digits in MNIST? I was thinking that a reasonable approach to answering the question would be to ...
100 views

### pytorch code for VAE for MINST forces mu and logvar to zero

I am new to pytorch and trying to implement a VAE for MNIST data. When I try to train my model, it appears that the model forces mu and logvar to zero (or something very close to zero) independent of ...
78 views

### Is vanishing KL a problem in vision based VAEs as well?

I came across some work on the problem of a vanishing KL contrbution in Variational Auto Encoders Cyclical Annealing Schedule: A Simple Approach to Mitigating KL Vanishing. This work particularly is ...
110 views

### Why KL Divergence instead of Cross-entropy in VAE

I understand how KL divergence provides us with a measure of how one probability distribution is different from a second, reference probability distribution. But why are they particularly used (...
43 views

### Varitional Autoencoder not accepting batch size or validation data

The input to the VAE will be a customer vector where the index of the vector represents a product id, position i in vector x is set to one iff product id i has been purchased by the customer. For ...
20 views

### Training a VAE for random number generation

I have a high-dimensional multi-modal distribution of random numbers in R^n and consider training a VAE to learn the distribution. What I want to do with it succeedingly, is to sample from the latent ...
206 views

### MSE loss in VAE reduces only KL divergence

I am trying to reproduce the two stage VAE from this repository in TF2 with Keras to learn MNIST digits. Unfortunately I am experiencing behavior that I can not explain to myself: As far as I ...
154 views

### What makes the posterior intractable?

In the setting of Variational AutoEncoders, i.e. when we want to find the posterior distribution over the data generating, latent variable z, given some ...
320 views

### VAE generates bad images. due to unbalanced loss functions?

I'm training a variational autoencoder on CelebA dataset using TensorFlow.keras The problem I'm facing is that the generated images are not diverse enough and look kinda bad. (new) Example: What I ...
288 views

### How to make custom callback in keras to generate sample image in VAE training?

I'm training a simple VAE model on 64*64 images and I would like to see the images generated after every epoch or every couple batches to see the progress. when I train the model I wait until the ...
403 views

### Train a GAN on “before and after” images of dental surgeries [closed]

I want a GAN to train on "before and after" images of dental surgeries; so that it can generate "after" pictures for fresh patients. Input images are like these: https://img.webmd.com/dtmcms/live/...
41 views

### Can VAEs be used to generate multivariate data?

Most of the tutorials online seem to use VAEs to generate images and use CNNs to generate data. I am working on a game with multivariate data consisting of character position and the character ...
191 views

### Variational Autoencoder: Negative log likelihood not optimized

I am using the auto encoding variational Bayes algorithm for one unsupervised object detection task. In the loss function, the reconstruction loss is calculated as the log likelihood of the original ...
88 views

### How to estimate total correlation KL[q(z)||Πjq(zj)] of VAE after training (useful for latents disentanglement evaluation)

FactorVAE and β-TCVAE both use total correlation (TC) batch estimation for their objectives. Where TC is: $$KL\bigl( q(z)||\prod\nolimits_{j} q(z_{j})\bigr)$$ both estimates are applied to $q(z|x)$...
77 views

### What is the meaning of “probability distribution of p(x)” of something uncountable?

I'm studying VAE and new to both of the neural network and the statistic. After some researches, I could understand the rough concept of VAE. But what makes me confused is, the meaning of probability ...
609 views

### Intractability in Variational Autoencoders

I'm having difficulty understanding when integrals are intractable in variational inference problems. In a variational autoencoder with observation $x$ and latent variable $z$ we want to maximize data ...
2k views

### Why maximize ELBO in the variational autoencoder?

For a variational autoencoder, we have that: $$\mathcal{L}(x,\theta,\phi) := \mathbb{E}_{z \sim q_\phi(z|x)}[\log p_{\theta}(x|z)] -KL[q_{\phi}(z|x) ||p(z)]$$ This is called the variational lower ...
2k views

### ValueError: Cannot convert a partially known TensorShape to a Tensor: (?, 256)

I'm working on a sequence to sequence approach using LSTM and a VAE with an attention mechanism. ...
3k views

### What do we visualize in showing a VAE latent space?

I am trying to wrap my head around VAE's and have trouble understanding what is being visualized when people make scatter plots of the latent space. I think I understand the bottleneck concept; we go ...