Concentration in parabolic Lotka-Volterra equations : an asymptotic-preserving scheme Helene Hivert

Duration: 52 mins 57 secs

Share this media item:

Embed this media item:

<iframe width="_width_" height="_height_" src="https://sms.cam.ac.uk/media/3890998/embed" frameborder="0" scrolling="no" allowfullscreen></iframe>

Choose size:

About this item

Available Formats

About this item

Description:	Helene Hivert (École Centrale de Lyon) 12/04/2022 Programme: FKT SemId: 34925


Created:	2022-04-20 12:14
Collection:	Frontiers in analysis of kinetic equations
Publisher:	Helene Hivert
Copyright:	Isaac Newton Institute
Language:	eng (English)


Abstract:	We consider a population structured in phenotypical trait, which influcences the adaptation of individuals to their environment. Each individual has the trait of his parent, up to small mutations. We consider the problem in a regime of long time and small mutations. The distribution of the population is expected to concentrate at some dominant traits. Dominant traits can also evolve in time, thanks to mutations. From a technical point of view, the concentration phenomenon is described thanks to a Hopf-Cole transform in the parabolic model. The asymptotic regime is a constrained Hamilton-Jacobi equation [G. Barles, B. Perthame, 2008 & G. Barles, S. Mirrahimi, B. Perthame, 2009]. The uniqueness of the solution of the limit equation has been adressed very recently [V. Calvez, K.-Y. Lam, 2020]. Because of the lack of regularity of the constraint, it can indeed have jumps, the numerical approximation of the constrained Hamilton-Jacobi equation must be treated with care. We propose an asymptotic preserving scheme for the problem transformed with Hopf-Cole. We show that it converges outside of the asymptotic regime, and that it enjoys stability properties in the transition to the asymptotic regime. Eventually, we show that the limit scheme is convergent for the constrained Hamilton-Jacobi equation.

Abstract:

We consider a population structured in phenotypical trait, which influcences the adaptation of individuals to their environment. Each individual has the trait of his parent, up to small mutations. We consider the problem in a regime of long time and small mutations. The distribution of the population is expected to concentrate at some dominant traits. Dominant traits can also evolve in time, thanks to mutations. From a technical point of view, the concentration phenomenon is described thanks to a Hopf-Cole transform in the parabolic model. The asymptotic regime is a constrained Hamilton-Jacobi equation [G. Barles, B. Perthame, 2008 & G. Barles, S. Mirrahimi, B. Perthame, 2009]. The uniqueness of the solution of the limit equation has been adressed very recently [V. Calvez, K.-Y. Lam, 2020]. Because of the lack of regularity of the constraint, it can indeed have jumps, the numerical approximation of the constrained Hamilton-Jacobi equation must be treated with care.
We propose an asymptotic preserving scheme for the problem transformed with Hopf-Cole. We show that it converges outside of the asymptotic regime, and that it enjoys stability properties in the transition to the asymptotic regime. Eventually, we show that the limit scheme is convergent for the constrained Hamilton-Jacobi equation.

Available Formats

Format	Quality	Bitrate	Size
MPEG-4 Video	1280x720	1.19 Mbits/sec	472.85 MB	View	Download
MPEG-4 Video	640x360	376.3 kbits/sec	145.94 MB	View	Download
WebM	1280x720	564.9 kbits/sec	219.15 MB	View	Download
WebM	640x360	228.16 kbits/sec	88.52 MB	View	Download
iPod Video	480x270	475.09 kbits/sec	184.25 MB	View	Download
MP3	44100 Hz	249.78 kbits/sec	96.96 MB	Listen	Download
Auto *	(Allows browser to choose a format it supports)