Settore scientifico disciplinare di riferimento | (FIS/03) |

Ateneo | Università degli Studi di ROMA "La Sapienza" |

Struttura di afferenza | Dipartimento di FISICA |

Recapiti | Elenco recapiti telefonici |

Andrea.Crisanti@uniroma1.it |

Full Name Andrea Crisanti

Citizenship Italian

E-mail andrea.crisanti@uniroma1.it

Spoken Languages Italian, English, French

Part II – Education

Type Year Institution Notes (Degree, Experience,…)

University graduation 1985 Università La Sapienza, Roma Laurea in Fisica cum laude

Post-graduate studies 1985/88 Racah Institute of Physics, The Hebrew University, Israel

PhD 1988 The Hebrew University, Israel PhD in Physics

Part III – Appointments

IIIA – Academic Appointments

11/1988 04/1989 Institut de Physique Théorique, EPFL, Lausanne, Scholarship

05/1989 02/1990 Institut de Physique Théorique, EPFL, Lausanne, Research Assistant

03/1990 03/1991 Institut de Physique Théorique, Université de Lausanne, Lausanne, Maitre Assistant

04/1991 09/1991 Dipartimento di Fisica, Università dell'Aquila, Postdoc

10/1991 10/2000 Dipartimento di Fisica, Universita la Sapienza, Ricercatore

11/2000 present Dipartimento di Fisica, Universita la Sapienza, Professore Associato

IIIB – Other Appointments

03/1994 03/1994 EPFL Hôtes Académiques

2003 2011 IPTh, CEA, Saclay Long term periods (one week / three months) as Hôtes Académiques

01/2012 05/2012 CNRS, LPTHE, Universite Pierre et Marie Curie, Paris Hôtes Académiques

2013 2015 CNRS, LPTHE, Universite Pierre et Marie Curie, Paris Long term periods (one week/one month) as Hôtes Académiques

Part IV – Teaching experience

1989/90 EPFL, Lausanne, Switzerland Troisième cycle de la physique en Suisse Romande

1990/91 Université de Lausanne, Switzerland Mécanique

1991/92-1999/00 Università La Sapienza Metodi Matematici della Fisica (esercitazioni)

1993/94 Università La Sapienza Metodi Asintotici e Sviluppi Perturbativi (dottorato)

1994/95-1995/96 Università La Sapienza Fisica Sperimentale I (Laurea in Chimica, esercitazioni)

1995/96 Università de l'Aquila Fisica Teorica

1996/97 Università de l'Aquila Metodi Matematici per la Fisica

1996/97 Università La Sapienza Fisica (Laurea in Scienze Biologiche, esercitazioni)

1995/96 Università La Sapienza Meccanica Statistica (dottorato)

1998/99-2003/04 Università La Sapienza Metodi computazionali ed analitici di base della fisica

2000/01 Università La Sapienza Fisica Generale (laurea in Chimica)

2001/02-2004/05 Università La Sapienza Laboratorio di Calcolo

2002/03-2007/08 Università La Sapienza Laboratorio di Fisica Computazionale I

2008/09-2010/11 Università La Sapienza Laboratorio di Fisica Computazionale II

2003/04 Università La Sapienza Metodi di approssimazione per la Fisica

2004/05-2012/13 Università La Sapienza Metodi Numerici per la Fisica

2013/14-2005/16 Università La Sapienza Metodi Computazionali per la Fisica

2012/13-2014/15 Università La Sapienza Fisica Generale II (laurea in Matematica)

2015/16 Università La Sapienza Meccanica Statistica

Part V - Administrative experience

Dipartimento di Fisica, Università la Sapienza, Department Responsible for: Courses on computational physics (2001/2002)

Dipartimento di Fisica, Università la Sapienza, Member of: Commissione Attività informativa e divulgazione scientica

Dipartimento di Fisica, Università la Sapienza, Member of: Giunta di Dipartimento (2 mandati)

Dipartimento di Fisica, Università la Sapienza Member of: Commissione per lo sviluppo dei servizi informatici della biblioteca ed integrazione

con il polo SBN

Dipartimento di Fisica, Università la Sapienza, Department Responsible for: Segreteria Didattica del Dipartimento di Fisica

Dipartimento di Fisica, Università la Sapienza, Department Responsible for: Laboratorio supporto e sviluppo informatico

Dipartimento di Fisica, Università la Sapienza, Member of: Commissione di manutenzione

Dipartimento di Fisica, Università la Sapienza, Department Responsible for: Servizi informatici e di rete.

Part VI – Research Activities

Topics Brief Description

*) Neural Networks Models

Study of the role of disorder and asymmetry in neural networks models.

The synaptic interactions between pairs of neurons in real brain are not

symmetric. Removing the constraint of symmetric interactions in neural

networks may lead to more efficient and robust systems. For example the

introduction of small asymmetry destroys the spin glass phase, which

correspond in terms of neural networks corresponds to a state of confusion.

*) High Dimensional Disordered Spin Systems

The properties of high dimensional (mean field) disordered spin systems are

investigated using both equilibrium statistical mechanical methods and the

relaxation dynamical approach based on Langevin equations. In the static,

equilibrium statistical mechanics, approach the properties of disordered spin

systems are typically studied using the so-called replica method to perform

the average over disorder. This, however, approach requires the nontrivial

limit of vanishing number of replicas. The dynamical approach makes use

of the functional methods based on the Martin-Siggia-Rose functional and

does not require the introduction of replicas. However, the complex nature

of the phase space of disordered systems leads to a nontrivial interplay

between the long time and the thermodynamic limits. The connection

between the static and dynamic descriptions is not trivial, even in the mean

field limit.

*) Random Matrices, Low Dimensional Disordered Systems, Chaotic Dynamical Systems.

Properties of products of random matrices and their use for the analysis of

chaotic dynamical and low dimensional disordered spin systems. The

evolution of a dynamical system can be described by a map in the phase

space (Poincare' map). The properties of the dynamics can thus be obtained

from that of products of (suitable) matrices. The evaluation of the partition

function of 1D or 2D spin systems can be translated to that of the maximum

Lyapunov exponent of product of (random) transfer matrices. The

connection with products of random matrices allows for the use of efficient

analytical and numerical methods in the study these systems. Chaos in

dynamical systems refers to the sensible dependence on initial condition:

trajectories starting from arbitrarily close initial condition differ of a finite

quantity in a finite time. The presence of chaos may affect the behavior of

systems in sensible way. For example in predicting their future state, or

when the system interacts with some external forcing. Chaos is not limited

to classical systems, however its definition in quantum systems requires

some care.

*) Equilibrium and Nonequilibrium Thermodynamics, Fluctuation relations

Equilibrium and nonequilibrium statistical mechanics description of slow

relaxing glassy systems. Glassy systems are characterized by very long,

well separated, characteristic relaxation time scales, leading to history

dependent behaviors, commonly referred to as aging. The presence of these

time scales reflects the complex structure of the free-energy landscape

where the dynamical evolution takes place. The properties of a glassy

system can then be obtained from the study of the topology of the free

energy landscape. In a dynamical approach the properties of a glassy

behavior can be characterized by the (non equilibrium) fluctuations of

macroscopic quantities, such as work or heat, (fluctuation relations) or by

the response of the system to external perturbations (generalized

fluctuation/dissipation relations).

*) Turbulence and Transport in Fluids

Turbulence and transport in turbulent fluids. Turbulent fluids have chaotic

behaviors whose spatial and temporal properties can be described

statistically, either using models or direct numerical integration of the

partial differential equations describing the evolution of the fluids (Navier-

Stokes equations). The turbulent regime of a fluid has direct consequence

on how other quantities (either passive or active) can be transported by the

fluid. The study of transport properties of fluids can be formulated in terms

of predictability.

*) Spherical p-spin Models for Glassy Behavior

The dynamical behavior of disordered spherical p-spin models is described

in the mean field limit by Mode-Coupling equations of the same form as

those describing glassy systems. Moreover they static and dynamical

properties can be obtained analytically. They thus represent simple

statistical mechanical models for the analysis of glassy behaviors.

*) Statistical Mechanical approach to Random Lasers

The statistical properties of laser light in homogeneous cavities can be

analyzed by mapping the laser dynamics to an ordered Hamiltonian system,

where the nonlinearities describes mode interactions. In Random Laser the

cavity is absent and light confinement is produced multiple scattering due to

the disorder of the medium. In this case the mapping of the random laser

dynamics is with a disordered Hamiltonian system described by coupled

random spherical p-spin models.

*) Monte Carlo, Real Space Renormalization Group, Disordered Systems

The critical behavior of spin systems can be studied using direct numerical

Monte Carlo methods. Renormalization group based on the Migdal–

Kadanoff bond removal approach is often considered a simple and valuable

tool to understand the critical behavior of complicated statistical mechanical

models avoiding direct numerical simulations. In presence of quenched

disorder, however, predictions obtained with the Migdal–Kadanoff bond

removal approach quite often fail to quantitatively and qualitatively

reproduce critical properties obtained in the mean-field approximation or by

numerical simulations in finite dimensions. To overcome these limitations

the renormalization procedure must be generalized to more structured

hierarchical lattice.

*) Gene expression noise

Infectious and mendelian diseases originate from a perturbation, which is

traceable in most cases. For each disease initiating events should be

identifiable. A similar approach, however, fails in multifactorial diseases

such as Multiple Sclerosis (MS). Decades of studies of heritable and

nonheritable causes of MS have unequivocally identified elements

associated with the disease. Nonetheless, these associations are neither

sufficient nor necessary for the development and prediction of MS. The key

heritable and nonheritable factors show up in peculiar way, with subtle

variations. The major nondeterministic component of phenotypic variations

is stochastic gene expression (gene expression noise) which, along with

time, may lead to the development of MS. Characterization and study of the

properties of gene expression noise is central for the prediction of the

occurrence of MS.

*) Stochastic Resonance

Stochastic resonance in biological systems. Noise is intrinsic to biological

systems, which have "learned" how to take extract useful work or, generally

speaking, information from it. Experimental studies have shown that noise

can enhance tactile and proprioceptive stimuli by a stochastic resonance

mechanism. These studies, however, focused on large-scale properties and

the underlining mechanisms are not yet well characterized. The study and

modeling of the effect of noise on sensor neurons in terms of stochastic

resonance may shed new light on the role of noise in biological systems.

*) Growth Kinetic Models

The passage from reversible equations of motion for a many-particle system

to a mesoscopic description, e.g., via a coarse-graining process, leads to

irreversible kinetic equations. These equations describe the (slow)

relaxation of the system towards the equilibrium state. The properties can

be studied either by direct numerical solution of the kinetic equation or

employing functional methods.

Part VII – Publication Summary

Papers [International] 124

Books [scientific] 1 Products of random Matrices in Statistical Physics, Springer Verlag (1993)

Books [teaching] 2 yaC-Primer: Yet Another - C Primer: Linguaggio (Parte I) e Applicazioni (Parte II), Aracne Editrice (1996)