deep-learning-course

Winter Semester 2018/2019 - Albert-Ludwigs-Universität Freiburg

Introduction

In this repository you will find different exercises for the fundamentals of Deep Learning, such as Multilayer Perceptrons, Optimization and Regularization techniques, or CNN's.

The MNIST dataset is located in the root of the repo. In order to train the different algorithms, you need to copy the dataset to the python file location.

Glossary

• Exercise 01: Eigenvalue Decomposition, Matrix Inversion, Norms and Orthogonal Matrices
• Exercise 02: Backpropagation and MLP Implementation using the MNIST dataset
• Exercise 03: Optimization and Regularization for the MNIST dataset
• Exercise 04: PyTorch and Convolutional Networks (CNN)
• Exercise 05: Recurrent Neural Networks (RNN)
• Exercise 06: Practical Methodology (Hyperparameter Optimization)
• Exercise 07: Autoenconders and GANs
• Exercise 08: Bayesian Neural Networks (BNN)
• Exercise 09: Review