The Beg Rohu Summer School of Physics
Learning with Machines, Physics and Minds
2-14 June 2025

E. Agliari: Information processing in Hebbian networks.
Y. Le Cun: Tba
J-R. King: AI and Neuroscience: in search of the laws of intelligence
S. Mallat: From score diffusion to renormalisation group models in physics
M. Mézard: Generative diffusion from a statistical physics viewpoint
A. Murugan: Learning without neurons in physical systems
Seminars: O. Dauchot, J. Kempe
Applications are open and can be made by filling the online application form. Students and postdocs must join a CV (uploaded in the application form) and have a recommendation letter directly sent by email to b e g r o h u @ p h y s . e n s . f r. Confirmed researchers are asked to join a brief CV.
The deadline for application is the 30th of March 2025. The selection of participants will take place mid-April. Applicants will be informed shortly after.
The registration fees are 1250 euros and cover all lodging expenses. Participants wishing to attend sailing courses will have to pay an additional participation fee of 150 euros.
The French National Sailing School (Ecole Nationale de Voile), located on the site of Beg Rohu, in Saint Pierre Quiberon, is devoted to the training of sailing teams at the national and regional levels and to the professional training of future sailing teachers. It also has a department of research and development.
Sailing lessons
Sailing lessons will take place in the afternoon from 1.30 pm to 5 pm. Several types of courses are planned: 8 meters sailing boat, catamaran, and windsurf. The French National Sailing school will provide the instructors and the boats. Participants wishing to attend sailing courses will have to pay a participation fee of 150 euros (other than the summer school tuition of 1250 euros).
Location
The school is accommodated at the French National Sailing School (Ecole Nationale de Voile), located on the site of Beg Rohu, in the village Saint Pierre Quiberon, on the peninsula of Quiberon in southern Brittany. The nearest cities are Quiberon, Auray and Vannes, all located in the district ("département" in French) of Morbihan.
How to get here
From Paris, TGV express trains take 3.5h to reach Auray. Then get on the local train or the bus to Quiberon, which makes a stop at Saint Pierre Quiberon close to Beg Rohu. We will post detailed instructions for the arrival and the departure after the selection of applicants.
Schedule
The school will start on Monday the 2nd of June (arrival on Sunday the 1st afternoon) and end after lunch on Saturday the 14th of June. There will be 3 lectures per day, except on Sunday the 8th of June that will be lectures-free. We will have access to several boats to make a day-long sailing trip.
A typical day schedule of the summer school will consist in 3 lectures per day, each one 1.5 hours long. There will be 2 lectures in the morning. The first one will start at 8.30 am. After lunch there will be sailing lessons (or free time for the ones not taking sailing lessons), followed by the third lecture at 5.45 pm. Some evenings, after dinner, there will be questions and answers sessions and seminars.
The detailed schedule will be provided at the beginning of the school.
The participants
PhD students, postdocs, young and senior researchers are all welcome to apply to the Beg Rohu school. A proportion of half PhD students and half young and confirmed researchers would be ideal for allowing both pedagogical lectures and dynamic interactions.
The lectures and the professors
The aim of the school is twofold: providing for the participants a solid and advanced knowledge of main topics in statistical physics and condensed matter, but also giving rise to interactions between confirmed researchers. Ideally, the outcome of each session will be a boost in the research subjects covered by the lectures both because of the excellent training and of the emergence of fruitful interactions between participants. To this aim, each lecture progressively goes from the basics to the current research edge. Lectures and professors are chosen so that the topic of the session is tackled from different points of view; questions and answers sessions are organized in order to foster interactions between students and lecturers.
Lecturers are chosen both for the important role they played in the development of the research subjects covered at the school and their pedagogical capability. Lessons will be all given on blackboard (except for showing experimental curves, movies, etc).
Topics and lectures of previous sessions
2024 - Concepts and Methods of Statistical Physics
L. Canet: The Functional Renormalisation Group and Non-Equilibrium Physics
W. Krauth: The Second Markov Chain Revolution
P. Sollich: Path Integrals in Non-Equilibrium Statistical Physics
R. Speicher: Free Probability, Free Cumulants and Random Matrices
M. Tarzia: Sparse Random Matrices: Cavity Approach and Anderson Localization
2023 - Statistical Physics of Complex Systems
F. Bouchet: Statistical Physics Applied to Climate Dynamics
Y. Fyodorov: SuSy Approach to Random Matrices: an Introduction
M. Mézard: Statistical Physics of Inferece
G. Schehr: Extreme Statistics of Strongly Correlated Variables
L. Zdeborovà: Statistical Physics of Computation
2022 - Out of Equilibrium Dynamics
E. Agoritsas: Driven Disordered Systems
B. Derrida: Systems of Particles in Non-Equilibrium Physics
A. Nahum: Out of Equilibrium Dynamics and Quantum Circuits
V. Vitelli: Hydrodynamics and Phase Transitions in Active Matter
2021 - Statistical Mechanics and Emergent Phenomena in Biology
N. Brunel: Models of Memory in Networks of Neurons
A. Cavagna: Collective Behaviour in Biological Systems
S. Ganguli: Complex Neural Dynamics and High-Dimensional Data Analysis
F. van Wijland: Methods for Non-Equilibrium Statistical Mechanics
2019 - Glasses, Jamming, and Slow Dynamics
L. Berthier: Glass transition and amorphous solids
A. Liu: The jamming transition and the marginally jammed state
E. Vanden-Eijnden: Stochastic dynamics, rare events and large deviations
F. Zamponi: Theory of the glass and jamming transitions in infinite dimensions
2018 - Deep Neural Networks and Statistical Physics
G. Ben Arous: Complexity of random landscapes
L. Bottou: Training Deep Networks with Stochastic Gradient Algorithms
J. Kurchan: Rugged phase-spaces and the dynamics within them
S. Mallat: Learning Physics with Multiscale Deep Neural Networks
M. Mézard: Statistical physics of inference
H. Sompolinsky: Statistical Mechanics of Deep Neural Networks
2017 - Out of Equilibrium Dynamics, Evolution and Genetics
L.F. Cugliandolo: Out of Equilibrium Dynamics of Complex Systems
D.S. Fisher: Evolutionary Dynamics of Large Populations
Y.V. Fyodorov: Counting Equilibria in Complex Systems via Random Matrices
D.R. Nelson: Population Genetics in Space and Time
2016 - Concepts and Methods of Statistical Physics
B. Derrida: Fisher KPP Equation and Applications
Y. Kafri: Statistics of Rare Events and Large Deviations
A. Montanari: Selected Topics in Machine Learning
S. Redner: Non-Equilibrium Statistical Physics
C. Wetterich: Non-Perturbative Renormalization Group
2015 - Statistical Physics, Biology, Inference and Networks
W. Bialek: Statistical Mechanics for Real Biological Networks
J.-P. Bouchaud: Random Matrix Theory and Big Data Cleaning
M. Lässig: Current Challenges in Statistical Genetics
C. Moore: Physics, Computation, Phase Transitions, and Networks
P. Vivo: Random Matrices - Theory and Practice
2014 - Nonequilibrium Statistical Mechanics and Active Matter
I. Giardina: Collective Behaviour in Animal Groups
C. Godrèche: Statistics of Persistence in Nonequilibrium Systems and Beyond
C. Jarzynski: Nonequilibrium Thermodynamics of Small Systems
J. Kurchan: Out of Equilibrium: Transient, Driven, Conditioned
S. Ramaswamy: Active Matter: Mechanics, Statistics, Hydrodynamics
2013 - Disordered Systems
B. Altshuler: Anderson Localization and Beyond
G. Ben Arous: Universal Features of Slow Dynamics in Random Media
S. Majumdar: Random Matrix Theory and its Applications
G. Parisi: Mean-Field theory of Glassy Systems and Beyond
2012 - Glass and Jamming Transitions
S. Nagel: Jamming and Granular Matter
G. Semerjian: Glassy Aspects of Optimization Problems
G. Szamel: Slow and Glassy Dynamics
G. Tarjus: An Overview of the Glass Transition
2011 - Statistical Physics and Complex Systems
J.-P. Bouchaud: Statistical Physics Approaches to Economics and Finance
P. Diaconis: The Mathematics of Mixing Things Up
M. Mézard: Information, Physics, and Computation
M. Vergassola: Statistical Physics for Biological Systems
2010 - Concepts and Methods of Statistical Mechanics
G. Biroli: Statistical Dynamics
J.-S. Caux: Integrable Models in Atomic and Condensed Matter Physics
B. Delamotte: Non-Perturbative Renormalization Group
W. Krauth: Statistical Mechanics: Algorithms and Computations
2009 - Quantum physics out of equilibrium
N. Andrei: Out of equilibrium quantum impurities and the Bethe Ansatz
L. F. Cugliandolo: Slow dynamics of quantum and classical systems
C. Kollath Quantum dynamics, cold atoms and numerical methods
A. J. Millis: Non-equilibrium impurity models and quantum phase transitions
O. Parcollet: Methods for quantum many body problems out of equilibrium
2008 - Manifolds in random media, random matrices and extreme value statistics
J.-P. Bouchaud: Rare events and extreme value statistics
P. Ferrari: Random matrices and related problems
P. Le Doussal: Pinning of elastic objects in random media
S. Majumdar: Extreme statistics for correlated variables
An atypical summer school
The Beg Rohu summer school was born in 1984, because of the passion of its organizer for both physics and sailing. These two worlds, which both require hard work, thinking and concentration, met on the peninsula of Quiberon where lectures on statistical mechanics were given during one month every summer. Along the years, the school acquired the nickname "Beg Rohu" because of its location at the French National Sailing School. This summer school took place until 1997.
In 2007, G. Biroli and A. Lefévre decided to put the Beg Rohu school back to life, with the help of its creator C. Godrèche, and with a slightly changed format, including lectures in English and a duration of two weeks only. The first session took place in 2008, from the 16th to the 28th of June.
Giulio Biroli
The Beg Rohu summer school,
LPENS
24 rue Lhomond
75005 Paris, FRANCE
Director : Giulio Biroli (LPENS)
Co-organizers : Chiara Cammarota (Sapienza Univ.), Laura Foini (IPhT-CEA) and Marco Tarzia (LPTMC-Sorbonne Univ.)
The Beg Rohu Summer School is run by the Physics Department of Ecole Normale Supérieure with IPhT CEA as a partner institution. The funding for the school is provided by these two institutions and by the "Fondation CFM pour la recherche".
