InfoGAN is a type of generative adversarial network that modifies the GAN objective to encourage it to learn interpretable and meaningful representations. This is done by maximizing the mutual information between a fixed small subset of the GAN’s noise variables and the observations.
Formally, InfoGAN is defined as a minimax game with a variational regularization of mutual information and the hyperparameter $\lambda$:
$$ \min_{G, Q}\max_{D}V_{INFOGAN}\left(D, G, Q\right) = V\left(D, G\right)  \lambda{L}_{I}\left(G, Q\right) $$
Where $Q$ is an auxiliary distribution that approximates the posterior $P\left(c\mid{x}\right)$  the probability of the latent code $c$ given the data $x$  and $L_{I}$ is the variational lower bound of the mutual information between the latent code and the observations.
In the practical implementation, there is another fullyconnected layer to output parameters for the conditional distribution $Q$ (negligible computation ontop of regular GAN structures). Q is represented with a softmax nonlinearity for a categorical latent code. For a continuous latent code, the authors assume a factored Gaussian.
Source: InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial NetsPaper  Code  Results  Date  Stars 

Task  Papers  Share 

Disentanglement  8  18.60% 
Image Generation  5  11.63% 
General Classification  3  6.98% 
Image Classification  2  4.65% 
Unsupervised Image Classification  2  4.65% 
Translation  2  4.65% 
Medical Image Generation  1  2.33% 
Clustering  1  2.33% 
Dimensionality Reduction  1  2.33% 
Component  Type 


Feedforward Network

Feedforward Networks  
Leaky ReLU

Activation Functions  
ReLU

Activation Functions  
Softmax

Output Functions 