BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:DAMTP Statistical Physics and Soft Matter Seminar
SUMMARY:Markovian approximation for Brownian particles dri
ven by coloured noise - Yongjoo Baek\, DAMTP
DTSTART;TZID=Europe/London:20190226T130000
DTEND;TZID=Europe/London:20190226T140000
UID:TALK116911AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/116911
DESCRIPTION:Self-propelled particles form a class of nonequili
brium systems with constant injection of energy on
a microscopic scale. Given sufficient time-scale
separation\, the dynamics of such particles can be
modelled as Brownian motion violating the fluctua
tion-dissipation relation\, namely Langevin dynami
cs with an instantaneous damping force and a Gauss
ian colored noise with rapidly decaying correlatio
ns. To make the model analytically tractable\, pre
vious studies have proposed various Markovian appr
oximation schemes which replace the coloured noise
with a multiplicative Gaussian white noise\; howe
ver\, these approaches are not systematic and may
restore equilibrium-like steady-state behaviours\,
failing to capture the nonequilibrium aspects of
the steady state. In this talk\, I present a syste
matic Markovian approximation\, which yields a Lan
gevin equation with a multiplicative non-Gaussian
white noise. The nonzero skewness of the noise is
shown to be essential for correctly predicting the
evolution of the probability distribution functio
n. The approach provides a convenient and reliable
method for predicting nonequilibrium currents\, f
orces\, and first-passage time statistics associat
ed with self-propelled particles.
LOCATION:MR11\, CMS
CONTACT:Etienne Fodor
END:VEVENT
END:VCALENDAR