diff options
Diffstat (limited to 'report')
| -rw-r--r-- | report/build/cw1_vz215_np1915.pdf | bin | 233980 -> 0 bytes | |||
| -rw-r--r-- | report/fig/mix.png | bin | 20008 -> 22116 bytes | |||
| -rw-r--r-- | report/paper.md | 43 |
3 files changed, 43 insertions, 0 deletions
diff --git a/report/build/cw1_vz215_np1915.pdf b/report/build/cw1_vz215_np1915.pdf Binary files differdeleted file mode 100644 index 3a6e8a5..0000000 --- a/report/build/cw1_vz215_np1915.pdf +++ /dev/null diff --git a/report/fig/mix.png b/report/fig/mix.png Binary files differindex 3cc03bc..d2c6967 100644 --- a/report/fig/mix.png +++ b/report/fig/mix.png diff --git a/report/paper.md b/report/paper.md index 77d3db7..b4a2a63 100644 --- a/report/paper.md +++ b/report/paper.md @@ -1,7 +1,50 @@ +# Introduction + +A Generative Adversarial Network is a system in which two blocks, discriminator and generator are competing in a "minmax game", +in which the objective of the two blocks is respectively maximization and minimization of the function presented below, +until an equilibrium is reached. During the weights update performed through the optimization process, the generator and discrimitaor are +updated in alternating cycles. + +$$ V (D,G) = E_{x~p_{data}(x)}[logD(x)] + E_{zp_z(z)}[log(1-D(G(z)))] $$ + +The issue with shallow architectures (**present the example we used for mode collapse**) can be ontain really fast training, +while producing overall good results. + +One of the main issues that raises from this kind of architectures is mode collapse. As the discriminator keeps getting +better, the generator tries to focus on one single class label to improve its loss. This issue can be observed in figure +\ref{fig:mode_collapse}, in which we can observe how after 200 thousand iterations, the output of the generator only represents few +of the labels originally fed to train the network. At that point the loss function of the generator starts getting worse as shown in figure +\ref{fig:vanilla_loss}. As we observe, G-D balance in not achieved as the discriminator loss almost reaches zero, while the generator loss keeps +increasing. + +\begin{figure} +\begin{center} +\includegraphics[width=24em]{fig/generic_gan_loss.png} +\caption{Shallow GAN D-G Loss} +\label{fig:vanilla_loss} +\end{center} +\end{figure} + +\begin{figure} +\begin{center} +\includegraphics[width=24em]{fig/generic_gan_mode_collapse.pdf} +\caption{Shallow GAN mode collapse} +\label{fig:mode_collapse} +\end{center} +\end{figure} + + # DCGAN ## DCGAN Architecture description +Insert connection of schematic. + +The typical structure of the generator for DCGAN consists of a sequential model in which the input is fed through a dense layer and upsampled. +The following block involves Convolution+Batch_normalization+Relu_activation. The output is then upsampled again and fed to another Convolution+Batch_Normalization+Relu_activation block. The final output is obtained through a Convolution+Tanh_activation layer. The depth of the convolutional layers decreases from input to output. + +The discriminator is designed through blocks that involve Convolution+Batch_Normalization+LeakyReLU_activation+Dropout. The depth of the convolutional layers increases from input to output. + ## Tests on MNIST Try some **different architectures, hyper-parameters**, and, if necessary, the aspects of **virtual batch |
