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:Isaac Newton Institute Seminar Series
SUMMARY:Tensor Methods for Parameter Estimation and Bifurc
ation Analysis of Stochastic Reaction Networks - S
huohao Liao (University of Oxford)
DTSTART;TZID=Europe/London:20160315T150000
DTEND;TZID=Europe/London:20160315T160000
UID:TALK64954AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/64954
DESCRIPTION:Intracellular networks of interacting bio-molecule
s carry out many essential functions in living cel
ls\, but the molecular events underlying the funct
ioning of such networks are ubiquitously random. S
tochastic modelling provides an indispensable tool
for understanding how cells control\, exploit and
tolerant the biological noise. A common challenge
of stochastic modelling is to calibrate a large n
umber of model parameters against the experimental
data. Another difficulty is to study how the beha
viours of a stochastic model depends on its parame
ters\, i.e. whether a change in model parameters c
an lead to a significant qualitative change in mod
el behaviours (bifurcation). One fundamental reaso
n for these challenges is that the existing comput
ational approaches are susceptible to the curse of
dimensionality\, i.e.\, the exponential growth in
memory and computational requirements in the dime
nsion (number of species and parameters). Herein\,
we have developed a tensor-based computational fr
amew \; ork to address this computational chal
lenge. It is based on recently proposed low-parame
tric\, separable tensor-structured representations
of classical matrices and vectors. The framework
covers the whole process from solving the underlyi
ng equations to automated parametric analysis of t
he stochastic models such that the high cost of wo
rking in high dimensions is avoided. One notable a
dvantage of the proposed approach lies in its abil
ity to capture all probabilistic information of st
ochastic models all over the parameter space into
one single tensor-formatted solution\, in a way th
at allows linear scaling of basic operations with
respect to the number of dimensions. Within such f
ramework\, the existing algorithms commonly used i
n the deterministic framework can be directly used
in stochastic models\, including parameter infere
nce\, robustness analysis\, sensitivity analysis\,
and stochastic bifurcation analysis.
LOCATION:Seminar Room 2\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR