Astrophysics and Relativity

Name	Category	Phone	Office
Silverwater Tal	Graduate Students
Stein Josef	Emeriti	02 6584 806	Popick 109
Tang Xiaping	Post-Docs
Teboul Odelia	Graduate Students
Thaler Irina	Post-Docs
Tuchman Yitzhak	Faculty	02 6584 490	Popick 119
Waldman Roni	Researchers	02 6584 461	Popick 115A
Wandel Amri	Senior Scientists	02 6586 860	Popick 123
Weissbein Amir	Graduate Students	02 6585 529	Kaplun Bldg.
Ziskin Shlomi	Post-Docs

Pages

Upcoming Events

Yesterday

Nonlinear Physics seminar: "Statistical mechanics of wide but finite artificial neural networks"
- Start time: 12:00pm
- End date: Wednesday, December 6th
- End time: 01:30pm
- Lecturer: Mr. Gadi Naveh (HUJI)Abstract: Over the past decade, Deep neural networks (DNNs) have brought significant advancements to various fields of science and technology. Key to their success is feature learning, enabling them to extract useful representations from raw data. However, DNNs remain challenging to interpret and lack a comprehensive analytical understanding. Recent studies revealed that DNNs in the infinite width limit are equivalent to Gaussian Processes (GPs), which are well-understoodanalytically but lack feature learning capabilities due to pre-determined features.In this thesis, we aim to bridge the gap between GPs and finite DNNs, using approaches inspired by theoretical physics. Employing statistical field theory, we treat the DNN output as a random object from a statistical ensemble. We performed extensive computer simulations of DNNs trained in both stylized and real-world settings. As for analytical methods, we first employ a perturbative approach to obtain the leading correction in 1/width to the GP limit. Subsequently, we develop a self-consistent mean-field type theory, providing a GP description of finite DNNs with a data-dependent kernel that embodies feature learning properties.
- View this event in Google Calendar
Position at Kyoto
- Start time: 02:00pm
- End date: Wednesday, December 6th
- End time: 03:00pm
- View this event in Google Calendar

Search form

Astrophysics and Relativity

Pages