The foundation of successful machine learning algorithms is a profound and sometimes complex mathematical framework. The better the underlying math is understood and the theoretical correctness of the algorithm is proven mathematically, the better usually the performance of the algorithms are. In particular, if you want to optimize the architecture, fine-tune the parameters or adapt the machine learning algorithms to your needs, you will need a good understanding of the underlying math.

An excellent introduction into the necessary math can be found in the book of Goodfellow, Bengio and Courville Deep Learning.

Content

Linear Algebra

Probability Theory and Statistics