As part of my MSc in Artificial Intelligence and Machine Learning, the Deep Learning module helped me understand how neural networks learn from data and how their different architectures suit different tasks.

I started with the foundations of Artificial Neural Networks (ANNs), covering neurons, activation functions, forward propagation and backpropagation. These topics explained what happens during training, as a network adjusts its weights to reduce prediction errors.

A major part of the module focused on Convolutional Neural Networks (CNNs) and computer vision. I explored how these networks learn image features and how they can be used for image classification, object detection and segmentation. Understanding the differences between these tasks helped me see why the choice of model matters.

The module also covered practical challenges, including overfitting, model evaluation and computational requirements. Choosing an architecture and adjusting hyperparameters meant considering both the task and the resources available.

This module helped me connect the mathematics behind neural networks with their practical use. It also gave me a clearer understanding of the decisions involved in training a model and assessing how well it performs on data it has not seen before.