-
Mean-field game approach to epidemic propagation on networks
Authors:
Louis Bremaud,
Olivier Giraud,
Denis Ullmo
Abstract:
We investigate an SIR model of epidemic propagation on networks in the context of mean-field games. In a real epidemic, individuals adjust their behavior depending on the epidemic level and the impact it might have on them in the future. These individual behaviors in turn affect the epidemic dynamics. Mean-field games are a framework in which these retroaction effects can be captured. We derive dy…
▽ More
We investigate an SIR model of epidemic propagation on networks in the context of mean-field games. In a real epidemic, individuals adjust their behavior depending on the epidemic level and the impact it might have on them in the future. These individual behaviors in turn affect the epidemic dynamics. Mean-field games are a framework in which these retroaction effects can be captured. We derive dynamical equations for the epidemic quantities in terms of individual contact rates, and via mean-field approximations we obtain the Nash equilibrium associated with the minimization of a certain cost function. We first consider homogeneous networks, where all individuals have the same number of neighbors, and discuss how the individual behaviors are modified when that number is varied. We then investigate the case of a realistic heterogeneous network based on real data from a social contact network. Our results allow to assess the potential of such an approach for epidemic mitigation in real-world implementations.
△ Less
Submitted 24 January, 2025;
originally announced January 2025.
-
Modelling vehicle and pedestrian collective dynamics: Challenges and advancements
Authors:
Cécile Appert-Rolland,
Alexandre Nicolas,
Armin Seyfried,
Antoine Tordeux,
Denis Ullmo
Abstract:
In our urbanised societies, the management and regulation of traffic and pedestrian flows is of considerable interest for public safety, economic development, and the conservation of the environment. However, modelling and controlling the collective dynamics of vehicles and pedestrians raises several challenges. Not only are the individual entities self-propelled and hard to describe, but their co…
▽ More
In our urbanised societies, the management and regulation of traffic and pedestrian flows is of considerable interest for public safety, economic development, and the conservation of the environment. However, modelling and controlling the collective dynamics of vehicles and pedestrians raises several challenges. Not only are the individual entities self-propelled and hard to describe, but their complex nonlinear physical and social interactions makes the multi-agent problem of crowd and traffic flow even more involved. In this chapter, we purport to review the suitability and limitations of classical modelling approaches through four examples of collective behaviour: stop-and-go waves in traffic flow, lane formation, long-term avoidance behaviour, and load balancing in pedestrian dynamics. While stop-and-go dynamics and lane formation can both be addressed by basic reactive models (at least to some extent), the latter two require anticipation and/or coordination at the level of the group. The results highlight the limitations of classical force-based models, but also the need for long-term anticipation mechanisms and multiscale modelling approaches. In response, we review new developments and modelling concepts.
△ Less
Submitted 30 October, 2024;
originally announced October 2024.
-
Mean Field Game Approach to Non-Pharmaceutical Interventions in a Social Structure model of Epidemics
Authors:
Louis Bremaud,
Olivier Giraud,
Denis Ullmo
Abstract:
The design of coherent and efficient policies to address infectious diseases and their consequences requires to model not only epidemics dynamics, but also individual behaviors, as the latter has a strong influence on the former. In our work, we provide a theoretical model for this problem, taking into account the social structure of a population. This model is based on a Mean Field Game version o…
▽ More
The design of coherent and efficient policies to address infectious diseases and their consequences requires to model not only epidemics dynamics, but also individual behaviors, as the latter has a strong influence on the former. In our work, we provide a theoretical model for this problem, taking into account the social structure of a population. This model is based on a Mean Field Game version of a SIR compartmental model, in which individuals are grouped by their age class and interact together in different settings. This social heterogeneity allows to reproduce realistic situations while remaining usable in practice. In our game theoretical approach, individuals can choose to limit their contacts by making a trade-off between the risks incurred by infection and the cost of being confined. The aggregation of all these individual choices and optimizations forms a Nash equilibrium through a system of coupled equations that we derive and solve numerically. The global cost born by the population within this scenario is then compared to its societal optimum counterpart (i.e. the optimal cost from the society viewpoint), and we investigate how the gap between these two costs can be partially bridged within a constrained Nash equilibrium for which a governmental institution would impose lockdowns. Finally we consider the consequences of the finiteness of the population size $N$, or of a time $T$ at which an external event would end the epidemic, and show that the variation of these parameters could lead to first order phase transitions in the choice of optimal strategies. In this paper, all the strategies considered to mitigate epidemics correspond to non-pharmaceutical interventions (NPI), and we provide here a theoretical framework within which guidelines for public policies depending on the characteristics of an epidemic and on the cost of restrictions on the society could be assessed.
△ Less
Submitted 12 April, 2024;
originally announced April 2024.
-
Mean-Field Games Modeling of Anticipation in Dense Crowds
Authors:
Matteo Butano,
Cécile Appert-Rolland,
Denis Ullmo
Abstract:
Understanding and modeling pedestrian dynamics in dense crowds is a complex yet essential aspect of crowd management and urban planning. In this work, we investigate the dynamics of a dense crowd crossed by a cylindrical intruder using a Mean-Field Game (MFG) model. By incorporating a discount factor to account for pedestrians' limited anticipation and information processing, we examine the model'…
▽ More
Understanding and modeling pedestrian dynamics in dense crowds is a complex yet essential aspect of crowd management and urban planning. In this work, we investigate the dynamics of a dense crowd crossed by a cylindrical intruder using a Mean-Field Game (MFG) model. By incorporating a discount factor to account for pedestrians' limited anticipation and information processing, we examine the model's ability to simulate two distinct experimental configurations: pedestrians facing the obstacle and pedestrians giving their back to the intruder. Through a comprehensive comparison with experimental data, we demonstrate that the MFG model effectively captures essential crowd behaviors, including anticipatory motion and collision avoidance.
△ Less
Submitted 28 May, 2024; v1 submitted 2 March, 2024;
originally announced March 2024.
-
Modeling of obstacle avoidance by a dense crowd as a Mean-Field Game
Authors:
Matteo Butano,
Thibault Bonnemain,
Cécile Appert-Rolland,
Alexandre Nicolas,
Denis Ullmo
Abstract:
In this paper we use a minimal model based on Mean-Field Games (a mathematical framework apt to describe situations where a large number of agents compete strategically) to simulate the scenario where a static dense human crowd is crossed by a cylindrical intruder. After a brief explanation of the mathematics behind it, we compare our model directly against the empirical data collected during a co…
▽ More
In this paper we use a minimal model based on Mean-Field Games (a mathematical framework apt to describe situations where a large number of agents compete strategically) to simulate the scenario where a static dense human crowd is crossed by a cylindrical intruder. After a brief explanation of the mathematics behind it, we compare our model directly against the empirical data collected during a controlled experiment replicating the aforementioned situation. We then summarize the features that make the model adhere so well to the experiment and clarify the anticipation time in this framework.
△ Less
Submitted 28 May, 2024; v1 submitted 1 March, 2024;
originally announced March 2024.
-
Epidemic models on homogeneous networks : some analytical results
Authors:
Louis Bremaud,
Olivier Giraud,
Denis Ullmo
Abstract:
The ability to actually implement epidemic models is a crucial stake for public institutions, as they may be overtaken by the increasing complexity of current models and sometimes tend to revert to less elaborate models such as the SIR. In our work, we study a simple epidemic propagation model, called SIR-$k$, which is based on a homogeneous network of degree $k$, where each individual has the sam…
▽ More
The ability to actually implement epidemic models is a crucial stake for public institutions, as they may be overtaken by the increasing complexity of current models and sometimes tend to revert to less elaborate models such as the SIR. In our work, we study a simple epidemic propagation model, called SIR-$k$, which is based on a homogeneous network of degree $k$, where each individual has the same number $k$ of neighbors. This model is more refined than the basic SIR which assumes a completely homogeneous population. We show that nevertheless, analytical expressions, simpler and richer than the ones existing for the SIR model, can be derived for this SIR-$k$ model. In particular we obtain an exact implicit analytical solution for any $k$, from which quantities such as the epidemic threshold or the total number of agents infected during the epidemic can be obtained. We furthermore obtain simple exact explicit solutions for small $k$'s, and in the large $k$ limit we find a new formulation of the analytical solution of the basic SIR model, which comes with new insights.
△ Less
Submitted 18 December, 2023;
originally announced December 2023.
-
Discounted Mean-Field Game model of a dense static crowd with variable information crossed by an intruder
Authors:
Matteo Butano,
Cécile Appert-Rolland,
Denis Ullmo
Abstract:
It was demonstrated in [Bonnemain et al., Phys. Rev. E 107, 024612 (2023)] that the anticipation pattern displayed by a dense crowd crossed by an intruder can be successfully described by a minimal Mean-Field Games model. However, experiments show that changes in the pedestrian knowledge significantly modify the dynamics of the crowd. Here, we show that the addition of a single parameter, the disc…
▽ More
It was demonstrated in [Bonnemain et al., Phys. Rev. E 107, 024612 (2023)] that the anticipation pattern displayed by a dense crowd crossed by an intruder can be successfully described by a minimal Mean-Field Games model. However, experiments show that changes in the pedestrian knowledge significantly modify the dynamics of the crowd. Here, we show that the addition of a single parameter, the discount factor $γ$, which gives a lower weight to events distant in time, is sufficient to observe the whole variety of behaviors observed in the experiments. We present a comparison between the discounted MFG and the experimental data, also providing new analytic results and insight about how the introduction of $γ$ modifies the model.
△ Less
Submitted 6 February, 2024; v1 submitted 17 February, 2023;
originally announced February 2023.
-
A social structure description of epidemics propagation with the mean field game paradigm
Authors:
Louis Bremaud,
Denis Ullmo
Abstract:
We consider the spread of infectious diseases through a Mean Field Game version of a SIR compartmental model with social structure, in which individuals are grouped by their age class and interact together in different settings. In our game theoretical approach, individuals can choose to limit their contacts if the epidemic is too virulent, but this effort comes with a social cost. We further comp…
▽ More
We consider the spread of infectious diseases through a Mean Field Game version of a SIR compartmental model with social structure, in which individuals are grouped by their age class and interact together in different settings. In our game theoretical approach, individuals can choose to limit their contacts if the epidemic is too virulent, but this effort comes with a social cost. We further compare the Nash equilibrium obtained in this way with the societal optimum that would be obtained if a benevolent central planner could decide of the strategy of each individual, as well as to the more realistic situation where an approximation of this optimum is reached through social policies such as lockdown.
△ Less
Submitted 22 June, 2022;
originally announced June 2022.
-
Pedestrians in static crowds are not grains, but game players
Authors:
Thibault Bonnemain,
Matteo Butano,
Théophile Bonnet,
Iñaki Echeverría-Huarte,
Antoine Seguin,
Alexandre Nicolas,
Cécile Appert-Rolland,
Denis Ullmo
Abstract:
The local navigation of pedestrians amid a crowd is generally believed to involve no anticipation beyond (at best) the avoidance of the most imminent collisions. We show that current models rooted in this belief fail to reproduce some key features experimentally evidenced when a dense static crowd is crossed by an intruder. We identify the missing ingredient as the pedestrians' ability to plan th…
▽ More
The local navigation of pedestrians amid a crowd is generally believed to involve no anticipation beyond (at best) the avoidance of the most imminent collisions. We show that current models rooted in this belief fail to reproduce some key features experimentally evidenced when a dense static crowd is crossed by an intruder. We identify the missing ingredient as the pedestrians' ability to plan their motion well beyond the next interaction, whence they may accept to move towards denser regions for a short time. To account for this effect, we introduce a minimal model based on mean-field game theory, which proves remarkably successful in replicating the aforementioned observations as well as other daily-life situations involving collective behaviour in dense crowds, such as partial metro boarding. This demonstrates the ability of game approaches to capture the anticipatory effects at play in operational crowd dynamics.
△ Less
Submitted 22 July, 2022; v1 submitted 21 January, 2022;
originally announced January 2022.
-
Lax Connection and Conserved Quantities of Quadratic Mean Field Games
Authors:
Thibault Bonnemain,
Thierry Gobron,
Denis Ullmo
Abstract:
Mean Field Game is a rather new field initially developed in applied mathematics and engineering in order to deal with the dynamics of a large number of controlled agents or objects in interaction. For a large class of these models, there exists a deep relationship between the associated system of equations and the non linear Schrödinger equation, which allows to get new insights on the structure…
▽ More
Mean Field Game is a rather new field initially developed in applied mathematics and engineering in order to deal with the dynamics of a large number of controlled agents or objects in interaction. For a large class of these models, there exists a deep relationship between the associated system of equations and the non linear Schrödinger equation, which allows to get new insights on the structure of their solutions. In this work, we deal with related aspects of integrability for such systems, exhibiting in some cases a full hierarchy of conserved quantities, and bringing some new questions which arise in this specific context.
△ Less
Submitted 1 December, 2020; v1 submitted 18 August, 2020;
originally announced August 2020.
-
Schrödinger approach to Mean Field Games with negative coordination
Authors:
Thibault Bonnemain,
Thierry Gobron,
Denis Ullmo
Abstract:
Mean Field Games provide a powerful framework to analyze the dynamics of a large number of controlled agents in interaction. Here we consider such systems when the interactions between agents result in a negative coordination and analyze the behavior of the associated system of coupled PDEs using the now well established correspondence with the non linear Schrödinger equation. We focus on the long…
▽ More
Mean Field Games provide a powerful framework to analyze the dynamics of a large number of controlled agents in interaction. Here we consider such systems when the interactions between agents result in a negative coordination and analyze the behavior of the associated system of coupled PDEs using the now well established correspondence with the non linear Schrödinger equation. We focus on the long optimization time limit and on configurations such that the game we consider goes through different regimes in which the relative importance of disorder, interactions between agents and external potential varies, which makes possible to get insights on the role of the forward-backward structure of the Mean Field Game equations in relation with the way these various regimes are connected.
△ Less
Submitted 28 July, 2020; v1 submitted 1 June, 2020;
originally announced June 2020.
-
Universal behavior in non stationary Mean Field Games
Authors:
Thibault Bonnemain,
Thierry Gobron,
Denis Ullmo
Abstract:
Mean Field Games provide a powerful framework to analyze the dynamics of a large number of controlled objects in interaction. Though these models are much simpler than the underlying differential games they describe in some limit, their behavior is still far from being fully understood. When the system is confined, a notion of "ergodic state" has been introduced that characterizes most of the dyna…
▽ More
Mean Field Games provide a powerful framework to analyze the dynamics of a large number of controlled objects in interaction. Though these models are much simpler than the underlying differential games they describe in some limit, their behavior is still far from being fully understood. When the system is confined, a notion of "ergodic state" has been introduced that characterizes most of the dynamics for long optimization times. Here we consider a class of models without such an ergodic state, and show the existence of a scaling solution that plays similar role. Its universality and scaling behavior can be inferred from a mapping to an electrostatic problem.
△ Less
Submitted 11 July, 2019;
originally announced July 2019.
-
Can a periodically driven particle resist laser cooling and noise?
Authors:
A. Maitra,
D. Leibfried,
D. Ullmo,
H. Landa
Abstract:
Studying a single atomic ion confined in a time-dependent periodic anharmonic potential, we find large amplitude trajectories stable for millions of oscillation periods in the presence of stochastic laser cooling. The competition between energy gain from the time-dependent drive and damping leads to the stabilization of such stochastic limit cycles. Instead of converging to the global minimum of t…
▽ More
Studying a single atomic ion confined in a time-dependent periodic anharmonic potential, we find large amplitude trajectories stable for millions of oscillation periods in the presence of stochastic laser cooling. The competition between energy gain from the time-dependent drive and damping leads to the stabilization of such stochastic limit cycles. Instead of converging to the global minimum of the averaged potential, the steady-state phase-space distribution develops multiple peaks in the regions of phase space where the frequency of the motion is close to a multiple of the periodic drive. Such distinct nonequilibrium behaviour can be observed in realistic radio-frequency traps with laser-cooled ions, suggesting that Paul traps offer a well-controlled test-bed for studying transport and dynamics of microscopically driven systems.
△ Less
Submitted 16 May, 2019; v1 submitted 3 October, 2018;
originally announced October 2018.
-
Far-from-equilibrium noise heating and laser cooling dynamics in radio-frequency Paul traps
Authors:
A. Maitra,
D. Leibfried,
D. Ullmo,
H. Landa
Abstract:
We study the stochastic dynamics of a particle in a periodically driven potential. For atomic ions trapped in radio-frequency Paul traps, noise heating and laser cooling typically act slowly in comparison with the unperturbed motion. These stochastic processes can be accounted for in terms of a probability distribution defined over the action variables, which would otherwise be conserved within th…
▽ More
We study the stochastic dynamics of a particle in a periodically driven potential. For atomic ions trapped in radio-frequency Paul traps, noise heating and laser cooling typically act slowly in comparison with the unperturbed motion. These stochastic processes can be accounted for in terms of a probability distribution defined over the action variables, which would otherwise be conserved within the regular regions of the Hamiltonian phase space. We present a semiclassical theory of low-saturation laser cooling applicable from the limit of low-amplitude motion to large-amplitude motion, accounting fully for the time-dependent and anharmonic trap. We employ our approach to a detailed study of the stochastic dynamics of a single ion, drawing general conclusions regarding the nonequilibrium dynamics of laser-cooled trapped ions. We predict a regime of anharmonic motion in which laser cooling becomes diffusive (i.e., it is equally likely to cool the ion as it is to heat it), and can also turn into effective heating. This implies that a high-energy ion could be easily lost from the trap despite being laser cooled; however, we find that this loss can be counteracted using a laser detuning much larger than Doppler detuning.
△ Less
Submitted 16 May, 2019; v1 submitted 23 August, 2018;
originally announced August 2018.
-
Mean Field Games in the weak noise limit : A WKB approach to the Fokker-Planck equation
Authors:
Thibault Bonnemain,
Denis Ullmo
Abstract:
Motivated by the study of a Mean Field Game toy model called the "seminar problem", we consider the Fokker-Planck equation in the small noise regime for a specific drift field. This gives us the opportunity to discuss the application to diffusion problem of the WKB approach "a la Maslov", making it possible to solve directly the time dependant problem in an especially transparent way.
Motivated by the study of a Mean Field Game toy model called the "seminar problem", we consider the Fokker-Planck equation in the small noise regime for a specific drift field. This gives us the opportunity to discuss the application to diffusion problem of the WKB approach "a la Maslov", making it possible to solve directly the time dependant problem in an especially transparent way.
△ Less
Submitted 9 April, 2018;
originally announced April 2018.
-
Phase-space study of surface-electrode Paul traps: Integrable, chaotic, and mixed motions
Authors:
V. Roberdel,
D. Leibfried,
D. Ullmo,
H. Landa
Abstract:
We present a comprehensive phase-space treatment of the motion of charged particles in electrodynamic traps. Focusing on five-wire surface-electrode Paul traps, we study the details of integrable and chaotic motion of a single ion. We introduce appropriate phase-space measures and give a universal characterization of the trap effectiveness as a function of the parameters. We rigorously derive the…
▽ More
We present a comprehensive phase-space treatment of the motion of charged particles in electrodynamic traps. Focusing on five-wire surface-electrode Paul traps, we study the details of integrable and chaotic motion of a single ion. We introduce appropriate phase-space measures and give a universal characterization of the trap effectiveness as a function of the parameters. We rigorously derive the commonly used (time-independent) pseudopotential approximation, quantify its regime of validity and analyze the mechanism of its breakdown within the time-dependent potential. The phase space approach that we develop gives a general framework for describing ion dynamics in a broad variety of surface Paul traps. To probe this framework experimentally, we propose and analyze, using numerical simulations, an experiment that can be realized with an existing four-wire trap. We predict a robust experimental signature of the existence of trapping pockets within a mixed regular and chaotic phase-space structure. Intricately rich escape dynamics suggest that surface traps give access to exploring microscopic Hamiltonian transport phenomena in phase space.
△ Less
Submitted 31 May, 2018; v1 submitted 5 April, 2018;
originally announced April 2018.
-
Quadratic Mean Field Games
Authors:
Denis Ullmo,
Igor Swiecicki,
Thierry Gobron
Abstract:
Mean field games were introduced independently by J-M. Lasry and P-L. Lions, and by M. Huang, R.P. Malhamé and P. E. Caines, in order to bring a new approach to optimization problems with a large number of interacting agents. The description of such models split in two parts, one describing the evolution of the density of players in some parameter space, the other the value of a cost functional ea…
▽ More
Mean field games were introduced independently by J-M. Lasry and P-L. Lions, and by M. Huang, R.P. Malhamé and P. E. Caines, in order to bring a new approach to optimization problems with a large number of interacting agents. The description of such models split in two parts, one describing the evolution of the density of players in some parameter space, the other the value of a cost functional each player tries to minimize for himself, anticipating on the rational behavior of the others.
Quadratic Mean Field Games form a particular class among these systems, in which the dynamics of each player is governed by a controlled Langevin equation with an associated cost functional quadratic in the control parameter. In such cases, there exists a deep relationship with the non-linear Schrödinger equation in imaginary time, connexion which lead to effective approximation schemes as well as a better understanding of the behavior of Mean Field Games.
The aim of this paper is to serve as an introduction to Quadratic Mean Field Games and their connexion with the non-linear Schrödinger equation, providing to physicists a good entry point into this new and exciting field.
△ Less
Submitted 10 August, 2018; v1 submitted 25 August, 2017;
originally announced August 2017.
-
An [imaginary time] Schrödinger approach to mean field games
Authors:
Igor Swiecicki,
Thierry Gobron,
Denis Ullmo
Abstract:
Mean Field Games (MFG) provide a theoretical frame to model socio-economic systems. In this letter, we study a particular class of MFG which shows strong analogies with the {\em non-linear Schrödinger and Gross-Pitaevski equations} introduced in physics to describe a variety of physical phenomena ranging from deep-water waves to interacting bosons. Using this bridge many results and techniques dev…
▽ More
Mean Field Games (MFG) provide a theoretical frame to model socio-economic systems. In this letter, we study a particular class of MFG which shows strong analogies with the {\em non-linear Schrödinger and Gross-Pitaevski equations} introduced in physics to describe a variety of physical phenomena ranging from deep-water waves to interacting bosons. Using this bridge many results and techniques developed along the years in the latter context can be transferred to the former. As an illustration, we study in some details an example in which the "players" in the mean field game are under a strong incentive to coordinate themselves.
△ Less
Submitted 1 December, 2015; v1 submitted 22 October, 2015;
originally announced November 2015.
-
'Phase diagram' of a mean field game
Authors:
Igor Swiecicki,
Thierry Gobron,
Denis Ullmo
Abstract:
Mean field games were introduced by J-M.Lasry and P-L. Lions in the mathematical community, and independently by M. Huang and co-workers in the engineering community, to deal with optimization problems when the number of agents becomes very large. In this article we study in detail a particular example called the 'seminar problem' introduced by O.Guéant, J-M Lasry, and P-L. Lions in 2010. This mod…
▽ More
Mean field games were introduced by J-M.Lasry and P-L. Lions in the mathematical community, and independently by M. Huang and co-workers in the engineering community, to deal with optimization problems when the number of agents becomes very large. In this article we study in detail a particular example called the 'seminar problem' introduced by O.Guéant, J-M Lasry, and P-L. Lions in 2010. This model contains the main ingredients of any mean field game but has the particular feature that all agent are coupled only through a simple random event (the seminar starting time) that they all contribute to form. In the mean field limit, this event becomes deterministic and its value can be fixed through a self consistent procedure. This allows for a rather thorough understanding of the solutions of the problem, through both exact results and a detailed analysis of various limiting regimes. For a sensible class of initial configurations, distinct behaviors can be associated to different domains in the parameter space . For this reason, the 'seminar problem' appears to be an interesting toy model on which both intuition and technical approaches can be tested as a preliminary study toward more complex mean field game models.
△ Less
Submitted 5 March, 2015;
originally announced March 2015.
