9 votes
Accepted

What is the meaning of $V(D,G)$ in the GAN objective function?

To understand this equation first you need to understand the context in which it is first introduced. We have two neural networks (i.e. $D$ and $G$) that are playing a minimax game. This means that ...
user avatar
  • 3,083
4 votes

How can we find find the input image which maximizes the class-probability for an ANN?

Probably the simplest way to search for an image with the highest probability of being a cat is to use a technique similar to Deep Dream: Load the network for training, but freeze all the network ...
user avatar
  • 23.2k
4 votes
Accepted

How can we find find the input image which maximizes the class-probability for an ANN?

In deep networks there is actually a wide variety of solutions to the problem, but if you need to find one, any easy way to do this is just through normal optimization schemes $$\hat x = argmin_x \ L(...
user avatar
  • 2,229
4 votes
Accepted

Other deep learning image generation techniques besides GANs?

There are several generative models that have been proposed before or roughly at the same time of the GAN (2014). For example, the deep Boltzman machine (2009), deep generative stochastic network (...
user avatar
  • 33k
4 votes

Why is the variational auto-encoder's output blurred, while GANs output is crisp and has sharp edges?

The key is: VAE usually use a small latent dimension, the information of input is so hard to pass through this bottleneck, meanwhile it tries to minimize the loss with the batch of input data, you ...
user avatar
4 votes
Accepted

Where is the mistake in my derivation of the GAN loss function?

I guess the issue is you lost track of where the samples came from and since you requested a math explanation I'll try to go step by step using my notation and without checking other material to avoid ...
user avatar
4 votes
Accepted

Would it be possible to determine the dataset a neural network was trained on?

You can already do this with some neural networks, such as GANs and VAEs, which are generative models that learn a probability distribution over the inputs, so they learn how to produce e.g. images ...
user avatar
  • 33k
3 votes
Accepted

What is the purpose of the GAN?

GANs were invented in a bar somewhere in Montreal, Canada. At said bar, the idea was that neural networks could be used for generating new examples from an existing distribution. This was the problem: ...
user avatar
3 votes

Book(s) on generative models

From the theoretical foundations one can look into the Chapter 20: Deep Generative Models of the classic DL book by Goodfellow, Bengio https://amzn.to/2MmZNbH. Not ...
user avatar
3 votes
Accepted

How to determine the quality of synthetic data?

Due to subjective nature, quantitative evaluation of synthetic images is difficult in general. However, there are metrics like Inception Score or FID score that are used for evaluation of generative ...
user avatar
  • 238
2 votes

How can we process the data from both the true distribution and the generator?

Let's start at the beginning. GANs are models that can learn to create data that is similar to the data that we give them. When training a generative model other than a GAN, the easiest loss function ...
user avatar
  • 1,144
2 votes

Neural network to get input attributes using only the output value

Now I have a network trained to get an output value from an random set of attributes but, can I use this trained network to get the input attributes using only the desired output? It depends: If you ...
user avatar
  • 23.2k
2 votes

Why do we need Upsampling and Downsampling in Progressive Growing of Gans

Use of Transposed Convolution can lead to checkerboard artifacts. So we prefer to up-sample and then apply convolution. You can check this article for more information https://distill.pub/2016/deconv-...
user avatar
2 votes

Are deep learning models suitable for training with sparse data?

The problem isn't the GAN but the implementation of its discriminator which is typically a convolutional neural network (CNN). CNNs have trouble with sparse data. They require dense data to learn ...
user avatar
2 votes
Accepted

Why do we use $D(x \mid y)$ and not $D(x,y)$ in conditional generative adversarial networks?

It looks like you're asking about the difference between using conditional and joint probabilities. The joint probability $$D(x,y)$$ is the probability of x and y both happening together. The ...
user avatar
2 votes
Accepted

How can AI be used to more reliably analyze and plan around the tie between climate and emissions?

Can AI provide a more reliable analysis of the gross effects of carbon emissions on extinctions of species ice-cap melting, and other effects? Yes. The work of Judea Pearl and others over the last 20 ...
user avatar
2 votes
Accepted

What are the possible social consequences of training neural networks with artificially generated data?

We can already observe information bubbles on social media, where the circle is that the ML algorithms learn what content people like and give more similar content based on clicks and so on. From a ...
user avatar
  • 927
2 votes
Accepted

Query regarding the minmax loss function formulation of the training of a Generative Adversarial Network (GAN)

I'll answer your questions one by one: In this equation are the $E_{z \sim p_z(z)}$ and $E_{x \sim p_{data}(x)}$ the means of the distributions of the mini batch samples? So let's take the first ...
user avatar
  • 3,083
2 votes

What parameters can be tweaked to avoid a generator or discriminator loss collapsing to zero when training a DC-GAN?

GANs are notably hard to train and it is not uncommon to have large bumps in the losses. The learning rate is a good start but the instability may come from a wide variety of reasons. I'm assuming ...
user avatar
  • 246
2 votes

Which approach can I use to generate text based on multiple inputs?

Generally, text generators work by modeling the joint distribution of the text by its Bayesian forward decomposition $ \begin{align*} p(w_1, w_2, ..., w_n) &= p(w_1) * p(w_2|w_1) * p(w_3|w_2, ...
user avatar
  • 2,229
2 votes
Accepted

Does MMD-VAE solve the problem of blurred images of vanilla VAEs?

[Answering my own question after 5 months of studying VAE models] The point of the MMD-VAE or InfoVAE is not exactly to emphasise on the visual quality of generated samples. It is to preserve greater ...
user avatar
  • 128
2 votes

Giving an AI a purpose to talk

I just came across this piece of news yesterday: "This week, Microsoft Research threw down the gauntlet with the launch of a competition challenging researchers around the world to develop AI agents ...
user avatar
2 votes

What is the state of the art in melody generation?

you do not need ai for that, just a little bit of math / statistics: audio: https://m.soundcloud.com/user-919775337/sets/algorithmic-reinterpretation method: https://stats.stackexchange.com/questions/...
user avatar
2 votes

What is the fundamental difference between the synthesis task and sampling task?

The terminologies can be confusing because of the different ways authors use them. The bottom line is this The Synthesis task basically refers to creating or synthesizing new data. Creation of data ...
user avatar
  • 238
2 votes
Accepted

What are the roles of the prior $\mathrm{p}(\mathbf{z})$ in a VAE?

The prior $p(z)$ is assumed as part of the problem formulation. A typical case is where $z$ is a vector of iid normal random variables. The ELBO involves a regularization term which encourages $q(z \, ...
user avatar
  • 746
1 vote
Accepted

What is the purpose of the noise injection in the generator network of a GAN?

Your goal is to model a distribution when constructing a GAN, therefore you need a way to be able to sample that distribution. The noise's purpose is so you can do this. Generally, it's drawn from a ...
user avatar
  • 2,229
1 vote

Which libraries can be used for image caption generation?

Image Caption Generation is an interesting problem to work on. I think your question was to know if there are any open-source libraries with built-in functions for Image Captioning. You can build ...
user avatar
1 vote

How important is it that the generator of a generative adversarial network doesn't take in information about input classes?

If you're building a straight "vanilla" generative adversarial network, it's best to understand the network as a statistical engine: You are training the generator on samples of a statistical ...
user avatar
  • 161
1 vote
Accepted

How to get good results with GAN and some thousands of images?

With a Google Cloud V100 GPU the GAN would run a week to two with default parameters. Does this sound realistic time for this kind of dataset? It's definitely not feasible for me. Yes, V100s are ...
user avatar

Only top scored, non community-wiki answers of a minimum length are eligible