The program has been composed by:
4th april 2018
5th april 2018
6th april 2018
Invited Speakers and their contributions
Jean-Philippe Bouchaud (École Polytechnique, Paris)
Teaching statistical methods for economics and social science
As P. W. Anderson wrote in 1972 in "More is different", the behavior of large assemblies of individuals
cannot be understood from the extrapolation of the behavior of isolated individuals.
On the contrary, completely new behaviors, sometimes spectacular and difficult to anticipate,
may appear and require new ideas and methods.
The object of statistical physics is precisely to try to understand these collective phenomena,
which do not belong to any of the underlying elementary constituents.
In particular, small changes at the individual level can lead to dramatic effects at the collective level.
Several simple examples will be discussed in the context of Social Sciences,
which demonstrate the need (and the difficulty) to go beyond classical models
in economics, based on the idea of a "representative agent".
Sergio Ciliberto (ENS, Lyon)
Teaching statistical physics through key experiments
In this lecture we will present several concepts of Statistical Physics which can be easily introduced
to students starting from experimental observations and data analysis. The goal is to show
that 1) the equilibrium statistical mechanics is useful to calibrate instruments such as optical traps
and atomic force microscopy 2) Non equilibrium statistical physis can be indeed useful to calibrate
force or to measure useful thermodynamic quantities in equilibrium and out of equilbrium. 3) the
relationship between information and thermodynamics. We first introduce the Brownian motion
and the electrical noise in a resistors, which have been the two first historical examples where fluctu-
ations dissipation theorem has been introduced. We will show movies of the motion of an ensemble
of Brownian particles and of a trapped colloid. We will then describe the measured noise on a
resistance and the power spectrum of an oscillator excited by its thermal noise. These experimental
results will allow the students to establish relationships between response function, noise spectral
density, Fluctution dissipation theorem, equipartion and probability density distribution. We will
also show how these concepts can be actually used to fully calibrate an experimenal apparatus. We
can then drive the systems with external forces and show simple exaples of driven systems such as
an electronic circuit and driven harmonic oscillator. In this way the concepts of fluctuating work
and heat can be introduced to explain in simple terms the physical meaning of Fluctuation Theorem
which is particularly explicit in the case of electronic circuits. In this case Seifert entropy can be
simply explained using a realistic example. We will discuss how these concept of out equilibrium
thermodynamics can be useful to calibrate small forces or small heat dissipation. Finally the con-
nections between information and thermodynamics can be illustrated using simple proof of principle
experiments on Brownian particles or single electorns transitors.
Eytan Domany (Weizmann Institute of Science, Rehovot )
Statistical physics and medical application
Raymond Kapral (University of Toronto, Toronto)
Statistical physics, condensed matter and chemistry
Synthetic nanomotors: a case study of how statistical physics methods, from microscopic to macroscopic,
can be used to understand their properties Synthetic chemically self-propelled nanomotors that move
in solution by a diffusiophoretic mechanism are active particles with interesting single-motor and collective properties.
The talk will show how the use of stochastic thermodynamic and microscopic approaches that incorporate microscopic
reversibility can be used to explore the dynamics and statistical properties of motor propulsion.
The collective behavior of many such motors is controlled by direct, chemical and hydrodynamic interactions,
and information on the nature of the many-motor dynamics provided by microscopic and stochastic approaches will be discussed.
Joachim Krug (Institut für Theoretische Physik, Köln)
Teaching statistical and biological physics at the University of Cologne
The M.Sc. program at the University of Cologne comprises several areas of specialization,
one of which is Statistical and Biological Physics. I will briefly outline the curriculum,
which is aimed at preparing students for research projects in experimental and theoretical biological physics
with a focus on evolutionary and genetic themes.
The lecture course on evolutionary biology and population genetics will be described in more detail,
and points of methodological and conceptual contact with (nonequilibrium) statistical physics will be emphasized.
Finally, some of the challenges associated with engaging students in interdisciplinary research
at the boundary between physics and biology will be discussed.
David Mukamel (Weizmann Institute of Science, Rehovot)
What is the role of models in teaching statistical physics?
Udo Seifert (Institut für Theoretische Physik, Stuttgart)
Teaching modern aspects of nonequilibrium statistical physics
I will discuss a minimal path to stochastic thermodynamics assuming familiarity with basic
classical thermodynamics and the basic models of simple diffusion.
One crucial element is the notion of a "time-reversed" process that leads easily to the fluctuation theorem(s)
and the Jarzynski and Crooks relation. The more recent thermodynamic uncertainty relation can be motivated
and introduced through its version for a thermodynamically consistent asymmetric random walk.
Applying it to experimental data of molecular motors can convince students of its striking power
as a tool for thermodynamic inference since it yields model-free bounds on the efficiency of such motors.
Angelo Vulpiani (Università di Roma I, Rome)
What is the role of first principles in teaching statistical physics?
The relevance of the probability theory for the statistical mechanics cannot be underestimated.
Beyond the mathematical aspects, at physical level a very important problem is:
what is the link between the probabilistic computations (i.e. the averages over an ensemble)
and the actual results obtained in laboratory experiments which, a fortiori,
are conducted on a single realization (or sample) of the system under investigation?
Student Speakers
Ido Borsini (ISAS University of Camerino)
Dissipative strctures
Francesco Cagnetta (School of Physics and Astronomy, University of Edinburgh)
The Kardar-Parisi-Zhang equation universality
The KPZ equation is a stochastic partial differential equation introduced in 1986 as a field theory of surface growth.
Since then, it has proven to be a topical equation in the field of nonequilibrium statistical physics,
in that it captures the large-scale features of problems as diverse as interacting random walkers
on a line and polymers in random media. Such a 'physical' universality blends with one of another sort,
given by the vast range of analytical tools developed and adopted in order to unveil the various facets of the equation.
In this talk I will give a student's perspective on the KPZ equation's worth as teaching/learning material,
focusing on how it is used in textbooks as a physical framework to introduce concepts
and mathematical techniques general to the whole field of statistical physics.
Salvatore Caruso (University of Modena and Reggio Emilia)
Kinetic Theory: a magnifying glass between scales of observation
Statistical Physics investigates large systems behaviour in terms of their microscopic costituents.
This approach leads to consider how different scales of observation affect the process of analysis.
In order to deal with such problems, Kinetic Theory represents a powerful tool;
it enables us to establish connections between microscopic and macroscopic realms.
In this talk I will stress its pedagogical relevance in physics programmes.
Furthermore, I will discuss the educational role of traditional examples,
toghether with more modern applications from research.
Chiara Franceschini (University of Modena and Reggio Emilia)
Stochastic Differential Equations
It turns out that, when considering random terms into a differential equation then one gets
a more realistic mathematical description of the situation.
For this reason it is natural to study stochastic differential equations.
In this talk I will focus on the prerequisites and difficulties (e.g. the fact that the derivative
of the Brownian motion is not well definied) a student would need and might have in order to define an SDE
as well as the background that students from physics and mathematics have in common and not.
Moreover, it would be interesting to see how SDE are introduced in two different books
(one with a physics approach via the brownian motion and one with a math approach
using the Ito integral and martingale).
If time allows we will solve some of the most well known SDE.
Leonardo Lenzini (University of Florence)
Statistical mechanics for long-range interacting systems: ensemble inequivalence
Emil Mallmin (University of Edinburgh)
Langevin, Fokker-Planck, and Master equations
Nonequilibrium physics covers a wide range of phenomena, using a large conceptual and mathematical toolbox.
In particular, there are three core modelling approaches: the Langevin, Fokker-Planck, and Master equations.
The new student of the subject faces the challenge of understanding not only each method separately,
but also their complicated interrelations. In this regard, I found Gardiner's classic "Stochastic Methods"
indispensable, with its emphasis on the Markov property as the unifying concept.
I will review how these modelling approaches are introduced in a few reference textbooks,
and offer my thoughts on the relative strength and weaknesses of different presentations, from the perspective of a student.
Alessandro Santini (University of Florence)
Linear theory of kinetic roughening: an undergradated point of view
Kinetic roughening was a subject of great interest during the eighties and nineties
of the last century. Many natural and mechanical processes lead to the
formation of rough interfaces. Examples of such phenomena are crystal
growth, vapor deposition and bacterial growth. In those years numerical
simulation, analytical theories and experiments achieved common results.
We are dealing with open systems far from equilibrium so the standard
techniques of statistical mechanics are not appropriate to describe the
problem of interface growth. However growing surfaces exhibit power
laws scaling behaviors and evolve to a steady state. This fact allows us
to describe roughness exponents and universality classes in a similar way of
the critical exponents in the proximity of a second-order phase transition.
I have recently studied this subject as an undergradueted
student who has not yet taken a class in equilibrium statistical mechanics.
I am here to talk about how it is possible to understand this topic and to
what extent for an outsider.
Clément Zankoc (University of Florence)
Models of active matter
Participants
Raffaella Burioni (Università di Parma)
Michele Campisi (Università di Firenze)
Duccio Fanelli (Università di Firenze)
Francesco Ginelli (University of Aberdeen)
Giuseppe Gonnella (Università degli Studi di Bari)
Stefano Lepri (CNR-ISC)
Arkady Pikovsky (Potsdam University)
Antonio Politi (University of Aberdeen)
Stefano Ruffo (SISSA)
Bernardo Spagnolo (Università di Palermo)