In Term 1 we plan to cover Functional Approximators (NN) & Dimensionality reduction methods.
For this section we mainly follow the book "Deep Learning architectures. A mathematical approach" by Ovidiu Calin
We meet every Wednesday 11:00-12:00 in B3.03 (Zeeman).
To get the latest updates, sign up to the mailing list.
| Date |
Topic |
Paper/Text book chapters |
Presenter |
| 12/10/22 |
Organisational meeting |
| 19/10/22 |
Introduction to Neural Networks |
"Deep Learning architectures. A mathematical approach" by Ovidiu Calin, Ch. 1, 5, 6
|
Claudia Viaro |
| 26/10/22 |
NN and approximation theorems |
"Deep Learning architectures. A mathematical approach" by Ovidiu Calin, Ch. 6, 7
|
Anna Kuchko |
| 02/11/22 |
Universal approximation theorems |
"Deep Learning architectures. A mathematical approach" by Ovidiu Calin, Ch. 9 |
Laszlo Udvardi |
| 09/11/22 |
Compression techniques and neural manifolds |
"Deep Learning architectures. A mathematical approach" by Ovidiu Calin, Ch. 13,14 |
Anna Kuchko |
| 16/11/22 |
Recurrent Neural Networks |
"Deep Learning architectures. A mathematical approach" by Ovidiu Calin, Ch. 17 |
Pablo Ramses Alonso Martin |
| Mid-term pub |
| 23/11/22 |
Graph Neural Networks |
Genaral theory+recent papers |
Richard Fox |
| 30/11/22 |
Physics-Informed Deep Learning |
"Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations" by M. Raissi, P. Perdikaris, G. E. Karniadakis |
Claudia Viaro |
| 7/12/22 |
Reservoir Computing & Echo State Networks |
"Echo state networks are universal" by L. Grigoryeva and J.-P. Ortega |
Guglielmo Gattiglio |
In AY 2022-2023 the reading group is organised by Claudia Viaro, Anna Kuchko and Laszlo Udvardi. Please don't hesitate to contact us if you have any questions or suggestions!
