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:BPI Seminar Series
SUMMARY:Automated parallel adjoints for model differentiat
ion\, optimisation and stability analysis - Patric
k Farrell\, Imperial College London
DTSTART;TZID=Europe/London:20131017T113000
DTEND;TZID=Europe/London:20131017T123000
UID:TALK47509AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/47509
DESCRIPTION:The derivatives of PDE models are key ingredients
in many important algorithms of computational scie
nce. They find applications in diverse areas such
as sensitivity analysis\, PDE-constrained optimisa
tion\, continuation and bifurcation analysis\, err
or estimation\, and generalised stability theory.\
n\nThese derivatives\, computed using the so-calle
d tangent linear and adjoint models\, have made an
enormous impact in certain scientific fields (suc
h as aeronautics\, meteorology\, and oceanography)
. However\, their use in other areas has been hamp
ered by the great practical difficulty of the deri
vation and implementation of tangent linear and ad
joint models. In his recent book\, Naumann (2011)
describes the problem of the robust automated deri
vation of parallel tangent linear and adjoint mode
ls as "one of the great open problems in the field
of high-performance scientific computingâ€ť.\n\nIn
this talk\, we present an elegant solution to this
problem for the common case where the original di
screte forward model may be written in variational
form\, and discuss some of its applications.
LOCATION:Open Plan Area\, BP Institute\, Madingley Rise CB3
0EZ
CONTACT:Catherine Pearson
END:VEVENT
END:VCALENDAR