-
Prof. Slava Rychkov (IHES)26/04/2022 09:00
Numerical simulations of Random Field Ising Model suggest that Parisi-Sourlas supersymmetry and dimensional reduction are present for spatial dimension d>d_c and are lost for d<d_c, where the critical dimension d_c is somewhere between 4 and 5. I will lecture about a recent theoretical framework, based on perturbative renormalization group, which predicts a mechanism for the loss of...
Aller à la page de la contribution -
Prof. Jorge Kurchan (ENS Paris)26/04/2022 11:00
The solution of the dynamics of glasses at the mean-field level - for example in the limit of high dimension -- has a few surprises. The equations possess an emergent reparametrization invariance. The supersymmetry associated with thermal equilibrium is broken, but a new one arises ‘from nowhere’. These two intriguing mathematical features dominate the glassy phenomenology.
Aller à la page de la contribution -
Prof. Nicholas Read (Yale University)26/04/2022 14:00
The mean field analysis of Ising spin glasses shows that at low temperature, these system display a non ergodic phase characterized by an exponential number of metastable states which is captured by the so called replica symmetry breaking (RSB) solution of the models.
Aller à la page de la contribution
However when this mean field theory picture is applied to finite dimensional systems, it runs into troubles. In these... -
Prof. Nicholas Read (Yale University)26/04/2022 15:30
The mean field analysis of Ising spin glasses shows that at low temperature, these system display a non ergodic phase characterized by an exponential number of metastable states which is captured by the so called replica symmetry breaking (RSB) solution of the models.
Aller à la page de la contribution
However when this mean field theory picture is applied to finite dimensional systems, it runs into troubles. In these... -
Prof. Ahmed El Alaoui (Cornell University)27/04/2022 09:00
These lectures will cover two main topics relating to some computational aspects of mean-field spin glass Gibbs measures:
1- Optimization: Can we efficiently find ground state configurations whenever they exist?
I will introduce the framework of Incremental Approximate Message Passing and the associated optimal control problem. This will be used to compute near optimal ground state...
Aller à la page de la contribution -
Prof. Slava Rychkov (IHES)27/04/2022 11:00
Numerical simulations of Random Field Ising Model suggest that Parisi-Sourlas supersymmetry and dimensional reduction are present for spatial dimension d>d_c and are lost for d<d_c, where the critical dimension d_c is somewhere between 4 and 5. I will lecture about a recent theoretical framework, based on perturbative renormalization group, which predicts a mechanism for the loss of...
Aller à la page de la contribution -
Prof. Jorge Kurchan (ENS Paris)27/04/2022 14:00
The solution of the dynamics of glasses at the mean-field level - for example in the limit of high dimension -- has a few surprises. The equations possess an emergent reparametrization invariance. The supersymmetry associated with thermal equilibrium is broken, but a new one arises ‘from nowhere’. These two intriguing mathematical features dominate the glassy phenomenology.
Aller à la page de la contribution -
Prof. Nicholas Read (Yale University)27/04/2022 15:30
The mean field analysis of Ising spin glasses shows that at low temperature, these system display a non ergodic phase characterized by an exponential number of metastable states which is captured by the so called replica symmetry breaking (RSB) solution of the models.
Aller à la page de la contribution
However when this mean field theory picture is applied to finite dimensional systems, it runs into troubles. In these... -
Prof. Jorge Kurchan (ENS Paris)28/04/2022 09:00
The solution of the dynamics of glasses at the mean-field level - for example in the limit of high dimension -- has a few surprises. The equations possess an emergent reparametrization invariance. The supersymmetry associated with thermal equilibrium is broken, but a new one arises ‘from nowhere’. These two intriguing mathematical features dominate the glassy phenomenology.
Aller à la page de la contribution -
Prof. Ahmed El Alaoui (Cornell University)28/04/2022 11:00
These lectures will cover two main topics relating to some computational aspects of mean-field spin glass Gibbs measures:
1- Optimization: Can we efficiently find ground state configurations whenever they exist?
I will introduce the framework of Incremental Approximate Message Passing and the associated optimal control problem. This will be used to compute near optimal ground state...
Aller à la page de la contribution -
Prof. Slava Rychkov (IHES)28/04/2022 14:00
Numerical simulations of Random Field Ising Model suggest that Parisi-Sourlas supersymmetry and dimensional reduction are present for spatial dimension d>d_c and are lost for d<d_c, where the critical dimension d_c is somewhere between 4 and 5. I will lecture about a recent theoretical framework, based on perturbative renormalization group, which predicts a mechanism for the loss of...
Aller à la page de la contribution -
Prof. Nicholas Read (Yale University)28/04/2022 15:30
The mean field analysis of Ising spin glasses shows that at low temperature, these system display a non ergodic phase characterized by an exponential number of metastable states which is captured by the so called replica symmetry breaking (RSB) solution of the models.
Aller à la page de la contribution
However when this mean field theory picture is applied to finite dimensional systems, it runs into troubles. In these... -
Prof. Ahmed El Alaoui (Cornell University)29/04/2022 09:00
These lectures will cover two main topics relating to some computational aspects of mean-field spin glass Gibbs measures:
1- Optimization: Can we efficiently find ground state configurations whenever they exist?
I will introduce the framework of Incremental Approximate Message Passing and the associated optimal control problem. This will be used to compute near optimal ground state...
Aller à la page de la contribution -
Prof. Slava Rychkov (IHES)29/04/2022 11:00
Numerical simulations of Random Field Ising Model suggest that Parisi-Sourlas supersymmetry and dimensional reduction are present for spatial dimension d>d_c and are lost for d<d_c, where the critical dimension d_c is somewhere between 4 and 5. I will lecture about a recent theoretical framework, based on perturbative renormalization group, which predicts a mechanism for the loss of...
Aller à la page de la contribution -
Prof. Ahmed El Alaoui (Cornell University)29/04/2022 14:00
These lectures will cover two main topics relating to some computational aspects of mean-field spin glass Gibbs measures:
1- Optimization: Can we efficiently find ground state configurations whenever they exist?
I will introduce the framework of Incremental Approximate Message Passing and the associated optimal control problem. This will be used to compute near optimal ground state...
Aller à la page de la contribution -
Prof. Jorge Kurchan (ENS Paris)29/04/2022 16:00
The solution of the dynamics of glasses at the mean-field level - for example in the limit of high dimension -- has a few surprises. The equations possess an emergent reparametrization invariance. The supersymmetry associated with thermal equilibrium is broken, but a new one arises ‘from nowhere’. These two intriguing mathematical features dominate the glassy phenomenology.
Aller à la page de la contribution -
Prof. Riccardo Zecchina (Bocconi University)30/04/2022 09:00
The scope of these lectures will be to discuss the statistical physics approach to the phase diagram and landscape of machine learning problems focusing on the loss landscape of artificial neural networks. We will show how this analysis suggest new powerful training algorithms based on the role of wide flat minima of the corresponding cost functions.
Aller à la page de la contribution -
Prof. Riccardo Zecchina (Bocconi University)30/04/2022 11:00
The scope of these lectures will be to discuss the statistical physics approach to the phase diagram and landscape of machine learning problems focusing on the loss landscape of artificial neural networks. We will show how this analysis suggest new powerful training algorithms based on the role of wide flat minima of the corresponding cost functions.
Aller à la page de la contribution -
Prof. Giorgio Parisi (Università di Roma La Sapienza)02/05/2022 09:00
-
Prof. Florent Krzakala (EPFL)02/05/2022 11:00
The scope of these lectures is to discuss the statistical physics approach to high dimensional inference and learning. In these problems one seeks for a particular configuration of some variables (the signal) which is hidden in a rough energy landscape of spurious non-informative minima. We will focus on two aspects of these problems: on the one hand we will show how statistical physics...
Aller à la page de la contribution -
Prof. Florent Krzakala (EPFL)02/05/2022 14:00
The scope of these lectures is to discuss the statistical physics approach to high dimensional inference and learning. In these problems one seeks for a particular configuration of some variables (the signal) which is hidden in a rough energy landscape of spurious non-informative minima. We will focus on two aspects of these problems: on the one hand we will show how statistical physics...
Aller à la page de la contribution -
Prof. Amin Coja-Oghlan (Goethe University Frankfurt)02/05/2022 15:30
Disordered models on Bethe lattices, i.e., the random d-regular graph, emerge naturally both in the context of the mean field treatment of spin glasses as well as in computer science where they arise as generic random instances of constraint satisfaction problems (CSPs).
Aller à la page de la contribution
Their analysis has been performed in the physics literature through the non-rigorous cavity method technique developed in... -
Prof. Gerard Ben Arous (New York University)03/05/2022 09:00
The energy landscape of mean field spin glasses displays a large (exponential in the dimension of the problem) number of minima and saddles. The geometrical properties of such random functions in high dimensions are of fundamental importance to understand the behavior of local algorithms that try to find optima in such landscapes (for example gradient descent).
Aller à la page de la contribution
The aim of these lectures we... -
Prof. Florent Krzakala (EPFL)03/05/2022 11:00
The scope of these lectures is to discuss the statistical physics approach to high dimensional inference and learning. In these problems one seeks for a particular configuration of some variables (the signal) which is hidden in a rough energy landscape of spurious non-informative minima. We will focus on two aspects of these problems: on the one hand we will show how statistical physics...
Aller à la page de la contribution -
Prof. Florent Krzakala (EPFL)03/05/2022 14:00
The scope of these lectures is to discuss the statistical physics approach to high dimensional inference and learning. In these problems one seeks for a particular configuration of some variables (the signal) which is hidden in a rough energy landscape of spurious non-informative minima. We will focus on two aspects of these problems: on the one hand we will show how statistical physics...
Aller à la page de la contribution -
Prof. Antonio Auffinger (Northwestern University)03/05/2022 15:30
Spin glasses at low temperature, display a non-ergodic phase characterized by an exponential number of pure states. The properties of the corresponding Gibbs measure have been characterized through the heuristic replica method culminated in the celebrated Parisi formula for low temperature spin glasses. The scope of these lectures will be to describe and analyze rigorously the properties of...
Aller à la page de la contribution -
Prof. Amin Coja-Oghlan (Goethe University Frankfurt)04/05/2022 09:00
Disordered models on Bethe lattices, i.e., the random d-regular graph, emerge naturally both in the context of the mean field treatment of spin glasses as well as in computer science where they arise as generic random instances of constraint satisfaction problems (CSPs).
Aller à la page de la contribution
Their analysis has been performed in the physics literature through the non-rigorous cavity method technique developed in... -
Prof. Gerard Ben Arous (New York University)04/05/2022 11:00
The energy landscape of mean field spin glasses displays a large (exponential in the dimension of the problem) number of minima and saddles. The geometrical properties of such random functions in high dimensions are of fundamental importance to understand the behavior of local algorithms that try to find optima in such landscapes (for example gradient descent).
Aller à la page de la contribution
The aim of these lectures we... -
Prof. Antonio Auffinger (Northwestern University)04/05/2022 14:00
Spin glasses at low temperature, display a non-ergodic phase characterized by an exponential number of pure states. The properties of the corresponding Gibbs measure have been characterized through the heuristic replica method culminated in the celebrated Parisi formula for low temperature spin glasses. The scope of these lectures will be to describe and analyze rigorously the properties of...
Aller à la page de la contribution -
Prof. Antonio Auffinger (Northwestern University)04/05/2022 15:30
Spin glasses at low temperature, display a non-ergodic phase characterized by an exponential number of pure states. The properties of the corresponding Gibbs measure have been characterized through the heuristic replica method culminated in the celebrated Parisi formula for low temperature spin glasses. The scope of these lectures will be to describe and analyze rigorously the properties of...
Aller à la page de la contribution -
Prof. Gerard Ben Arous (New York University)05/05/2022 09:00
The energy landscape of mean field spin glasses displays a large (exponential in the dimension of the problem) number of minima and saddles. The geometrical properties of such random functions in high dimensions are of fundamental importance to understand the behavior of local algorithms that try to find optima in such landscapes (for example gradient descent).
Aller à la page de la contribution
The aim of these lectures we... -
Prof. Amin Coja-Oghlan (Goethe University Frankfurt)05/05/2022 11:00
Disordered models on Bethe lattices, i.e., the random d-regular graph, emerge naturally both in the context of the mean field treatment of spin glasses as well as in computer science where they arise as generic random instances of constraint satisfaction problems (CSPs).
Aller à la page de la contribution
Their analysis has been performed in the physics literature through the non-rigorous cavity method technique developed in... -
Prof. Gerard Ben Arous (New York University)05/05/2022 14:00
The energy landscape of mean field spin glasses displays a large (exponential in the dimension of the problem) number of minima and saddles. The geometrical properties of such random functions in high dimensions are of fundamental importance to understand the behavior of local algorithms that try to find optima in such landscapes (for example gradient descent).
Aller à la page de la contribution
The aim of these lectures we... -
Prof. Antonio Auffinger (Northwestern University)05/05/2022 15:30
Spin glasses at low temperature, display a non-ergodic phase characterized by an exponential number of pure states. The properties of the corresponding Gibbs measure have been characterized through the heuristic replica method culminated in the celebrated Parisi formula for low temperature spin glasses. The scope of these lectures will be to describe and analyze rigorously the properties of...
Aller à la page de la contribution -
Prof. Amin Coja-Oghlan (Goethe University Frankfurt)06/05/2022 09:00
Disordered models on Bethe lattices, i.e., the random d-regular graph, emerge naturally both in the context of the mean field treatment of spin glasses as well as in computer science where they arise as generic random instances of constraint satisfaction problems (CSPs).
Aller à la page de la contribution
Their analysis has been performed in the physics literature through the non-rigorous cavity method technique developed in... -
Prof. Silvio Franz (Université Paris-Saclay, LPTMS)06/05/2022 11:00
Choisissez le fuseau horaire
Le fuseau horaire de votre profil: