In this article, the author shares insights from their journey of learning how to utilize deep learning in computer graphics, aiming to help others quickly grasp terms and concepts in deep learning. They introduce neural networks, particularly multilayer perceptron (MLP), as powerful tools for tasks like classification and regression. The article includes code and a sample application, highlighting the architecture and activation functions of neural networks. Training a neural network involves adjusting weights through gradient descent, optimizing the network for better performance. The author emphasizes the importance of choosing the right network size and learning rate for effective training. The article provides a comprehensive guide to implementing and using MLP neural networks.
https://gpuopen.com/learn/deep_learning_crash_course/