Written Assignments
| Building blocks and architecture of neural networks |
|
| Matlab’s Deep Learning Toolbox |
|
| A simple network to recognise handwritten digits |
|
| How the backpropagation algorithm works |
|
| Universality of neural networks |
|
| Training of deep networks - unstable and vanishing gradients |
|
| Deep learning : convolutional networks |
Talks
Talks take place Wednesdays 15:15-16:45, SR 210 (Brüderstraße 16)First meeting on April 3rd 15:15, SR 210
Matlab installation guide by Jasper Petschat: Installation of MATLAB
| Caleb Watson Building blocks and architecture of neural networks |
Slides | ||
| Jasper Petschat Matlab’s Deep Learning Toolbox |
Slides | Matlab Code | |
| Thomas Lettau A simple network to recognise handwritten digits |
Slides | Matlab Code | |
| Daniil Kudrin How the backpropagation algorithm works |
Slides | ||
| Heinrich-Gregor Zirnstein Improving the way neural networks learn - cross-entropy cost function, overfitting and regularization |
Notes | ||
| Denis Gessert Universality of neural networks |
Slides | Animations | |
| Markus Heber Training of deep networks - unstable and vanishing gradients |
Slides | Matlab Code | |
| Roman Worschech Deep learning : convolutional networks |
Slides | Matlab Code |
Course Information
| Neural Networks and Deep Learning | |
| Deep Learning, Ian Goodfellow and Yoshua Bengio and Aaron Courville | |
| Matlab Deep Learning Toolbox Documentation | |
| Matlab Deep Learning Toolbox |
