People Mail List

Name	Primary Category	Discipline
Adiel Etty	Administration
Agam Oded	Faculty	Condensed Matter Physics
Agranov Tal	Graduate Students	Nonlinear
Aharon Nati	Post-Docs	Condensed Matter Physics
Alpern Hen	Graduate Students	Condensed Matter Physics
Amusia Miron Ya	Emeriti	Condensed Matter Physics
Anahory Yonathan	Faculty	Condensed Matter Physics
Arad-Vosk Neta	Graduate Students	Condensed Matter Physics
Armon Tsafrir	Graduate Students	Nonlinear
Arviv Galya	Graduate Students

Pages

Upcoming Events

Tomorrow

Condensed-Matter Physics Seminar: "Magneto-enhancement of superconductivity in composite D-wave superconductors"
- Start time: 12:00pm
- End date: Thursday, June 21st
- End time: 01:30pm
- Lecturer: B.Z.Spivak, University of Washington Abstract:We study composite D-wave superconductors consisting of randomly oriented and randomly distributed superconducting droplets embedded into a matrix. In a certain range of parameters the application of a small magnetic field enhances the superconductivity in these materials while larger fields suppress superconductivity as usual in conventional superconductors. We investigate the magnetic field dependence of the superfluid density and the critical temperature of such superconductors.
- View this event in Google Calendar

Monday, June 25th

Physics Colloquium: "The Information Theory of Deep Neural Networks: The statistical physics aspects"
- Start time: 12:00pm
- End date: Monday, June 25th
- End time: 01:30pm
- Lecturer: Naftali Tishby, Hebrew University of Jerusalem Abstract:The surprising success of learning with deep neural networks poses two fundamental challenges: understanding why these networks work so well and what this success tells us about the nature of intelligence and our biological brain. Our recent Information Theory of Deep Learning shows that large deep networks achieve the optimal tradeoff between training size and accuracy, and that this optimality is achieved through the noise in the learning process. In this talk, I will focus on the statistical physics aspects of our theory and the interaction between the stochastic dynamics of the training algorithm (Stochastic Gradient Descent) and the phase structure of the Information Bottleneck problem. Specifically, I will describe the connections between the phase transition and the final location and representation of the hidden layers, and the role of these phase transitions in determining the weights of the network.Based partly on joint works with Ravid Shwartz-Ziv, Noga Zaslavsky, and Shlomi Agmon.
- View this event in Google Calendar

Tuesday, June 26th

astrolunch: Anatoli Fedynitch
- Start time: 12:30pm
- End date: Tuesday, June 26th
- End time: 01:30pm
- Hadronic interactions of Ultra-High Energy Cosmic Rays in the Earth’s atmosphere
  
  Abstract: While the major source of Ultra-High Energy Cosmic Rays (UHECR) is not yet identified, it is clear that only a few cosmic particle accelerators are potentially able to reach the required maximal energies of almost 10^21 eV. At these energies UHECR mostly interact with low-energy photons in the source and during propagation. The experimental evidence for a (light) nuclear CR mass composition at Earth complicates the situation. The disintegration of nuclei during transport imposes increased modeling complexity and requires uncertain inputs from nuclear physics, introducing degeneracies in the interpretation of the spectrum in terms of source parameters. Another, probably more severe, problem arises from the modeling of the Extensive Air Showers (EAS) that UHECR generate when interacting with the Earth’s atmosphere. The interpretation of the mass composition is almost entirely model dependent, where the highest uncertainty originates from the modeling of hadronic nucleus-air interactions. The current state-of-the-art event generators for this task are EPOS-LHC, Sibyll 2.3c, QGSJet II-04 and DPMJET-III. At present, however, none of the models succeeds in describing the number of muons that would be expected from the observed electromagnetic component. This implies that data taken with the surface detectors can not be exploited for composition sensitive analyses. In this seminar I will outline the principles of hadronic interaction modeling discussing in higher depth the simplest model – SIBYLL - providing some insight into conceptual and practical differences to the other models. I will also discuss current air shower observations and indirect constraints to review the present situation in this field.
- View this event in Google Calendar

Search form

People Mail List

Pages