Introduction to Deep Learning
April 26th, 2018 PIOTR MIGDAL Data Science Consultant, Warsaw

Deep learning is a name for machine learning techniques using many-layered artificial neural networks. Their progress in the last years was spectacular -reaching human-level accuracy or better in image recognition (for photos, medical data and satellite images), mastering Go and advancing natural language processing. And surely there is more to come!

I will provide an overview of deep learning applications and give insight into their structure. For a physicists neural networks are non-linear tensor expressions, penalized by cross-entropy and optimized by gradient descent. I will show how to use neural network in Python, in either Kereasor PyTorch(on day 2 I encourage you to bring your laptop).

To get some taste, see:

ICFO+ Theory Lecture, April 25-26, 2018, 10:30. Blue Lecture Room