Mirco Huennfeld: Deep Learning for Event Reconstruction
→
Europe/Brussels
Large Seminar Room (Universe)
Large Seminar Room
Universe
Description
Presenter: Mirco Huennfeld
Title: Deep Learning for Event Reconstruction
Abstract:
Recent advances, especially in image recognition, have shown the capabilities of deep learning. Deep neural networks can be extremely powerful and their usage is computationally inexpensive once the networks are trained. While the main bottleneck for deep neural networks in the traditional domain of image classification is the lack of sufficient labeled data, this usually does not apply to physics where millions of Monte Carlo simulations exist.
The IceCube Neutrino Observatory is a Cherenkov detector deep in the Antarctic ice where the reconstruction of muon-neutrino events is one of the key challenges. Due to limited computational resources and the high data rate, only simplified reconstructions limited to a small subset of data can be run on-site at the South Pole. However, in order to perform online analysis and to issue real-time alerts, a fast and powerful reconstruction is necessary. An approach using deep learning techniques can reduce the computational complexity while improving the resolution. I will give a very brief overview of the basic concepts of deep learning and present how these methods can be used in physics experiments such as IceCube.
