Questions tagged [autoencoders]

Question 1

I was curious if anyone happens to know why data augmentations (like color jitter, random cropping, etc) appear to not be always used when training autoencoder or neural-based compressors for images ...

Question 2

I am trying to use a convolutional autoencoder to perform dimensionality reduction with the ultimate goal of reconstructing temperature fields. As I understand it, the goal of an autoencoder is to ...

Question 3

I get that the decoder computes $p_\theta(x\mid z)$. Here $\theta$ is close to the true MLE which is intractable. But still $p_\theta(\cdot\mid z)$ is a probability distribution, and the decoder ...

Question 4

Are VAEs considered explainable AI? To me, they are because the latent variables are interpretable, e.g, you change one and you might see its effects on the head rotation (for a dataset of faces, for ...

Question 5

I have trouble understanding the minimization of the KL divergence. In this link https://www.ibm.com/think/topics/variational-autoencoder They say, "One obstacle to using KL divergence for ...

Question 6

Setup The variational autoencoder (VAE) loss is given by the following (see here, for example): $$L = - \sum_{j = 1}^J \frac{1}{2} \left(1 + \log (\sigma_i^2) - \sigma_i^2 - \mu_i^2 \right) - \frac{1}{...

Question 7

I am reading this paper https://openaccess.thecvf.com/content/CVPR2021/papers/Jaques_NewtonianVAE_Proportional_Control_and_Goal_Identification_From_Pixels_via_Physical_CVPR_2021_paper.pdf Are the ...

Question 8

I am quite new to ML and I am developing my first Variational AutoEncoder (VAE), which is composed of a CNN encoder (4 layers) and a CNN decoder (4 layers). The input images are of size 128x128 and ...

Question 9

I have developed a VAE to understand if it is able to distinguish lung images of COVID-19, Normal images, or images with Viral Pneumonia. The VAE is composed of CNN encoder and CNN decoder (shown ...

Question 10

I am looking into the relationship between linear Variational Autoencoder (VAE) and probabilistic PCA (pPCA) presented by Lucas et al. (2019). Don't blame the elbo! paper In the official ...

Question 11

As mentioned in the title, I understand the mathematical derivation of equations (6-7) in Kingma's original paper. \begin{equation} \log p_\theta(\mathbf{x}, y) \geq \mathbb{E}_{q_\phi(\mathbf{z} \mid ...

Question 12

Say you have 4 stacked output vectors of 4 different VAEs: $B \times 512 \times 4$ These $512$ elements correspond to $256 \ \mu$ & $256 \ \ln\sigma^2$ (log-variances) of four multi-variate ...

Question 13

I’m learning different approaches to impute a tabular dataset of mixed continuous and categorical variables, and with data assumed to be missing completely at random. I converted the categorical data ...

Question 14

I'm trying to train an LSTM Variational Autoencoder, but I cannot figure out why the model is not making any progress, the loss gets stuck immediately. Here is my code and training loop. The sequences ...

Question 15

Let's imagine I have time series data for 50 users and 20 features per user: User1_ts(F1, ...F20), User2_ts(F1, ...F20), ...User50_ts(F1, ... F20). F20 is my target variable for estimation, and each ...

Stack Exchange Network

Questions tagged [autoencoders]

Why aren't Data Augmentations used when training autoencoder or neural-based compressors for images / video?

Should Feature Maps be considered for Dimensionality Reduction with Convolutional AutoEncoders

Which $x$ is actually computed by the decoder in VAEs, for a given $z$?

Are VAEs considered explainable AI?

Simple question about VAEs

Backpropagating regularization term in variational autoencoders

Evidence lower bound in Newtonian VAEs

Is this overfit?

How to improve segregation of images in VAEs?

Linear VAE and pPCA

Why is the target 𝑦 used as an input to the encoder in a semi-supervised VAE model?

Weighted Fusion of VAE Gaussian Distributions

Intuition behind generative methods for imputing missing data

LSTM Variational auto encoder not learning with Pytorch [duplicate]

Apply PCA to multi individual time series dataset

Hot Network Questions