Convolutional Neural Network

A Convolutional Neural Network (CNN) is a deep neural network commonly used for analyzing visual information. It's particularly effective for tasks like image recognition, classification, and segmentation.

CNNs are fundamental for image generation, and we will use them to implement both Autoencoder and GAN architectures.

CNN architecture

The fundamental principle of a convolutional neural network (CNN) lies in its utilization of sliding windows that scan various parts of an image, extracting valuable features such as color, shape, and contours.

This process generates a collection of feature maps, each capturing different aspects of the input image. These feature maps are then utilized for tasks such as classification or latent representation of the image, enabling the network to learn and discern intricate patterns and structures within the data.

CNN Layers

Now let's consider the most important layers of CNN architecture.

Convolutional Layers

• Feature Extraction: convolutional layers apply filters (also called kernels) to the input data. These filters slide over the input data, performing element-wise multiplication between the filter weights and the corresponding pixels in the input image. This operation generates feature maps that highlight important patterns and structures in the data;

• Parameter Sharing: one key characteristic of convolutional layers is parameter sharing. Instead of learning separate parameters for every location in the input image, the same set of weights is used across the entire image. This reduces the number of parameters in the model, making it more efficient and capable of learning spatial hierarchies of features.

Pooling Layers

• Downsampling: pooling layers downsample the feature maps generated by convolutional layers. They do this by reducing the spatial dimensions of the feature maps while retaining the most important information. This helps in reducing computational complexity and preventing overfitting;

• Translation Invariance: pooling layers introduce translation invariance, meaning that the network becomes less sensitive to small translations or shifts in the input data. This property makes the network more robust to variations in input images.

Dense Layer

• The final dense layer is utilized to flatten the outputs of preceding layers and extract the final feature vector;
• This layer serves to aggregate the learned representations from earlier layers into a concise feature vector, enabling the network to make predictions or generate data based on these extracted features.

CNN vs MLP