BEGIN:VCALENDAR
PRODID:-//Ben Fortuna//iCal4j 1.0//EN
VERSION:2.0
CALSCALE:GREGORIAN
X-WR-CALNAME:Seminarium\, Optimeringslära och systemteori
BEGIN:VEVENT
DTSTAMP:20200918T095935Z
SUMMARY:Gianpiero Canessa: Static risk averse models and applications
DESCRIPTION:Abstract:\nStochastic optimization allows the modeler to inco
rporate risk into his decision making process. Different applications of
this optimization methodology will be presented\, showing in each case
the innovation incorporated into a classical deterministic model and tra
nsforming it into the stochastic version.\n\nThe main challenge of stoch
astic optimization is that one often generates models that cannot be sol
ved directly: we need to transform it to a tractable deterministic equiv
alent problem (DEP) and solve it using any of the commercial solvers ava
ilable. Naive reformulations into DEPs can\, and often will\, result in
complex and/or large DEPs that current solvers may not be able to solve
in an adequate amount of time\, or even load in memory due to its size.
This work is centered in showing how we can circumvent these difficultie
s and obtain results that are equivalent to those that would be obtained
in the original stochastic formulation.\n
LOCATION:F11
DTSTART:20190913T090000Z
DTEND:20190913T100000Z
UID:eccf7bd5-d8d6-43d9-9dff-26e82ca85b37
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20200918T095935Z
SUMMARY:Topics in Workforce management\, in a contact center context.
DESCRIPTION:Pre-defense Seminar\nGöran Svensson will give a preview of hi
s thesis that will be defended on the 27th of September
LOCATION:F11
DTSTART:20190920T090000Z
DTEND:20190920T100000Z
UID:94aea730-2933-4ac6-8d69-96a1e45fa6bc
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20200918T095935Z
SUMMARY:Shen Peng: Chance constrained problem and some applications.
LOCATION:F11
DTSTART:20191004T090000Z
DTEND:20191004T100000Z
UID:23e6fb34-4f42-4560-80c3-ade2adb59880
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20200918T095935Z
SUMMARY:Bin Zhu: An Empirical Bayes Approach to Frequency Estimation
DESCRIPTION:Abstract: In this talk\, we show that the classical problem o
f frequency estimation can be formulated and solved efficiently in an em
pirical Bayesian framework by assigning a uniform a priori probability d
istribution to the unknown frequency. We discover that the covariance ma
trix of the signal model is the discrete-time counterpart of the operato
r whose eigenfunctions are the famous prolate spheroidal wave functions\
, introduced by Slepian and coworkers in the 1960's and widely studied i
n the signal processing literature although motivated by a different cla
ss of problems. The special structure of the covariance matrix is exploi
ted to design an estimator for the hyperparameters of the prior distribu
tion which is essentially linear\, based on subspace identification. Thi
s is in contrast to standard parametric estimation methods which are bas
ed on iterative optimization algorithms of local nature. Simulations sho
w that the approach is quite promising and seems to compare very favorab
ly with classical methods from the literature.\n\nShort Bio: Bin Zhu was
born in Changshu\, Jiangsu Province\, China in 1991. He received the B.
Eng.~degree from Xi'an Jiaotong University\, Xi'an\, China in 2012 and t
he M.Eng.~degree from Shanghai Jiao Tong University\, Shanghai\, China i
n 2015\, both in control science and engineering. In 2019\, he obtained
a Ph.D. degree in information engineering from University of Padova\, Pa
dova\, Italy\, and now he is a postdoctoral researcher in the same unive
rsity. His current research interest includes spectral estimation by rat
ional covariance extension and Bayesian frequency estimation.\n
LOCATION:F11
DTSTART:20191011T090000Z
DTEND:20191011T100000Z
UID:0a50a2a5-4c71-4a12-839f-4e490f996938
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20200918T095935Z
SUMMARY:Alexander Aurell: Topics in the mean-field type approach to pedes
trian crowd modeling and conventions.
LOCATION:F11
DTSTART:20191206T100000Z
DTEND:20191206T110000Z
UID:160f5ecb-5a6a-4f26-adca-e81642f43959
END:VEVENT
END:VCALENDAR