A Machine Learning Compendium

At the time I was studying Microelectronics and Computer Science I had the opportunity to take some fascinating machine learning lectures. Since that time I have followed this topic up to now where the research in neural networks and machine learning gains a lot of momentum.

Please see the navigation menu for details on machine learning and neural networks, their usage for regression, prediction, and classification problems, for reinforcement learning, generative models and other interesting fields of application.

My interest in machine learning research and the growing amount of available publications led me to the decision to - once again - dive deeper into this topic.

To understand all the things down to their details, I decided to implement all components of neural networks including the optimization environment from scratch using Python and the NumPy library.

So I started based on my knowledge about neural networks as they were the days I studied, combining it with the latest research outcomes regarding new activation functions, new optimizer algorithms and new network structures, altogether better suited to solve several problems.

As a nice side effect I got a better understanding of Python, NumPy, and PyTorch.