BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Optimal convergence rates for sparsity promoting wavelet-regulariz
ation in Besov spaces
DTSTART;VALUE=DATE-TIME:20191107T150000Z
DTEND;VALUE=DATE-TIME:20191107T154000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-28@conference2.aau.at
DESCRIPTION:Speakers: Thorsten Hohage ()\nWe report on joint work with Phi
lip Miller on Tikhonov regularization for linear and nonlinear ill-posed o
perator equations with wavelet Besov norm penalties. We focus on Besov nor
ms with fine index 1\, which yield estimators that are sparse with respect
to a wavelet frame. Our framework includes\, among others\, the Radon tra
nsform and some nonlinear inverse problems in differential equations with
distributed measurements.\nUsing variational source conditions we show tha
t such estimators achieve minimax-optimal rates of convergence for finitel
y smoothing operators in certain Besov balls both for deterministic and fo
r statistical noise models.\nUsing explicit bounds on the number of nonvan
ishing coefficients of the estimator\, we also derive a converse result fo
r approximation rates.\n\nhttps://conference2.aau.at/event/16/contribution
s/28/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/28/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Brief introduction into the work as an actuary
DTSTART;VALUE=DATE-TIME:20191108T133000Z
DTEND;VALUE=DATE-TIME:20191108T134500Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-54@conference2.aau.at
DESCRIPTION:Speakers: Jonas Offtermatt ()\nIn this talk a short introducti
on into to work of a mathematician at an insurance company will be given.
Starting with the one and only necessary equation for insurance mathematic
s\, over working as data analyst\, or in marketing\, we will end with the
work of a mathematician in IT-department.\n\nhttps://conference2.aau.at/ev
ent/16/contributions/54/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/54/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Modelling approaches to assess the reliability of semiconductor de
vices
DTSTART;VALUE=DATE-TIME:20191108T123000Z
DTEND;VALUE=DATE-TIME:20191108T124500Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-29@conference2.aau.at
DESCRIPTION:Speakers: Olivia Pfeiler ()\nOne research area of KAI – Komp
etenzzentrum Automobil- und Industrieelektronik is the reliability of semi
conductor devices. Often\, the reliability is assessed by extensive testin
g. To save test resources\, we develop lifetime models describing the degr
adation of the devices under application conditions. To achieve good resul
ts we research three approaches: statistical lifetime models\, FEM simulat
ion models and stochastic state-space models.\n\nhttps://conference2.aau.a
t/event/16/contributions/29/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/29/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Minimization based formulations of the EIT problem with the CEM
DTSTART;VALUE=DATE-TIME:20191108T111500Z
DTEND;VALUE=DATE-TIME:20191108T113000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-31@conference2.aau.at
DESCRIPTION:Speakers: Kha Van Huynh ()\nOne of the recent approaches to so
lving inverse problems is to use the all-at-once formulation where both th
e state and the parameter are considered as unknowns. The advantage of thi
s method is to avoid using the parameter-to-state map\, which is usually d
ifficult to determine in real problems and sometimes leads to strictly res
trictive conditions. In this talk\, we regularize the electrical impedance
tomography (EIT) problem with the complete electrode model (CEM) in the p
lane. The regularization method is to formulate our problem as a minimizat
ion problem\, which is a generalization of the all-at-once formulation. Th
e new one is to use CEM\, a widely used practical model.\n\nhttps://confer
ence2.aau.at/event/16/contributions/31/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/31/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Iterative regularization for nonsmooth inverse problems
DTSTART;VALUE=DATE-TIME:20191107T085000Z
DTEND;VALUE=DATE-TIME:20191107T093000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-39@conference2.aau.at
DESCRIPTION:Speakers: Christian Clason ()\nWe consider inverse problems fo
r nonlinear forward models that are directionally but not Fréchet differe
ntiable\; examples include solution mappings for nonsmooth partial differe
ntial equations or variational inequalities. In this setting\, standard de
rivative-based regularization methods such as Landweber or Levenberg--Marq
uardt iteration are inapplicable. We show that using elements of the Bouli
gand subdifferential for the linearization still leads to a convergent reg
ularization scheme.\n\nhttps://conference2.aau.at/event/16/contributions/3
9/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/39/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Acceleration of sequential subspace optimization in Banach spaces
by orthogonal search directions
DTSTART;VALUE=DATE-TIME:20191107T140000Z
DTEND;VALUE=DATE-TIME:20191107T144000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-40@conference2.aau.at
DESCRIPTION:Speakers: Thomas Schuster ()\nA standard solution technique fo
r linear operator equations of first kind is the Landweber scheme which is
an iterative method that uses the negative gradient of the current residu
al as search direction\, which is then also called the Landweber direction
. Though this method proves to be stable with respect to noisy data\, it i
s known to be numerically slow for problems in Hilbert spaces and this beh
avior shows to be even worse in some Banach space settings. This is why th
e idea came up to use several search directions instead of the Landweber d
irection only which has led to the development of Sequential Subspace Opti
mization (SESOP) methods. This idea is related to Conjugate Gradient (CG)
techniques that are known to be amongst the most effective methods to solv
e linear equations in Hilbert spaces. SESOP methods in Banach spaces do no
t share the conjugacy property with CG methods. In this talk we present th
e concept of generalized (g-)orthogonality in Banach spaces and apply metr
ic projections to orthogonalize the current Landweber direction with respe
ct to the search space of the last iteration. This leads to an accelerated
SESOP method which is confirmed by various numerical experiments.\n(joine
d work with Frederik Heber and Frank Schoepfer)\n\nhttps://conference2.aau
.at/event/16/contributions/40/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/40/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Modelling and Machine Learning for industrial sensing applications
DTSTART;VALUE=DATE-TIME:20191108T131500Z
DTEND;VALUE=DATE-TIME:20191108T133000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-33@conference2.aau.at
DESCRIPTION:Speakers: Romana Boiger ()\nThe Materials Center Leoben (MCL)
is an internationally active research institution specialized in materials
\, production and processing engineering and innovative material applicati
ons. Besides a comprehensive range of services\, the MCL carries out coop
erative research and development projects in close collaboration with indu
stry. The latter is the focus of my team\, the embedded computing group: w
e support research projects along the whole product value chain from mater
ial design\, to material processing and finally their use in innovative pr
oducts. In this talk\, I show two recent examples of my work. This include
s a gas sensor device for indoor air quality control and algorithms for vi
sual anomaly detection in semiconductor industry. The mathematics therein
involve data analysis\, mathematical modelling of dynamical systems\, opti
mization\, inverse problems\, numerical methods and machine learning using
real data from industrial sensors.\n\nhttps://conference2.aau.at/event/16
/contributions/33/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/33/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Theoretical and numerical analysis of fundamental models in nonlin
ear acoustics
DTSTART;VALUE=DATE-TIME:20191107T095000Z
DTEND;VALUE=DATE-TIME:20191107T103000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-34@conference2.aau.at
DESCRIPTION:Speakers: Mechthild Thalhammer ()\nIn this talk\, I will prese
nt joint work with Barbara Kaltenbacher on fundamental models in nonlinear
acoustics. In a first part\, I will introduce a hierarchy of nonlinear da
mped wave equations that arise in the modelling of sound propagation in th
ermoviscous fluids. In a second part\, I will discuss a rigorous result wh
ich implies that two classical models\, the Kuznetsov and Westervelt equat
ions\, are retained as limiting systems for vanishing thermal conductivity
and consistent initial data. In addition\, I will present numerical illus
trations that complement the theoretical findings.\n\nhttps://conference2.
aau.at/event/16/contributions/34/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/34/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A high-order discontinuous Galerkin approach for nonlinear sound w
aves
DTSTART;VALUE=DATE-TIME:20191107T103000Z
DTEND;VALUE=DATE-TIME:20191107T110000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-35@conference2.aau.at
DESCRIPTION:Speakers: Vanja Nikolić ()\nAccurate simulation of nonlinear
sound waves offers a road to improving a variety of procedures in industry
and medicine\, \nranging from non-destructive detection of material damag
es to non-invasive treatments of medical disorders.\nIn this talk\, we wil
l discuss the spatial discretization of Westervelt’s quasi-linear acoust
ic equation by a high-order discontinuous Galerkin method. \nThe challenge
s in the numerical analysis lie in handling the nonlinearity in the model
which involves the derivatives in time of the acoustic velocity potential\
,\nand in preventing the model from degenerating. Numerical experiments wi
ll illustrate the theoretical convergence results.\nThis is joint work wit
h Paola F. Antonietti\, Ilario Mazzieri (MOX\, Politecnico di Milano)\, Ma
rkus Muhr\, and Barbara Wohlmuth (TU Munich).\n\nhttps://conference2.aau.a
t/event/16/contributions/35/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/35/
END:VEVENT
BEGIN:VEVENT
SUMMARY:All-at-once formulations of inverse problems within the Bayesian a
pproach
DTSTART;VALUE=DATE-TIME:20191108T110000Z
DTEND;VALUE=DATE-TIME:20191108T111500Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-36@conference2.aau.at
DESCRIPTION:Speakers: Anna Schlintl ()\nThe all-at-once-formulation for de
terministic inverse problems has recently been considered. Our goal is to
put this approach in a Bayesian framework. The advantages of our approach
are the additional choice of a prior also for the state variable and the p
ossibility to take into account perturbations in the model. By means of th
e inverse source problem and the backward heat equation we test the all-at
-once formulation in appropriate function spaces\, derive adjoint operator
s\, investigate in different priors for the state and do numerical tests.\
n\nhttps://conference2.aau.at/event/16/contributions/36/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/36/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Process Simulation at Linde Engineering
DTSTART;VALUE=DATE-TIME:20191108T130000Z
DTEND;VALUE=DATE-TIME:20191108T131500Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-37@conference2.aau.at
DESCRIPTION:Speakers: Stephania Hokenmaier ()\nBuilding on its vast and un
ique experience\, Linde has been developing and optimizing gas processing\
, separation and liquefaction technologies for 140 years. Through trusted\
, lasting business relationships\, it collaborates closely with customers
the world over to develop tailored solutions that maximize plant lifecycle
productivity\, efficiency and service life.\nWith the help of process sim
ulation tools entire chemical plants or parts of them are modelled\, simul
ated and also optimized. At Linde Engineering commercial and in-house deve
loped process simulation programs are applied. In this talk the idea\, met
hods and application of process simulation will be explained.\n\nhttps://c
onference2.aau.at/event/16/contributions/37/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/37/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Stochastic state space modeling of thermo-mechanical fatigue
DTSTART;VALUE=DATE-TIME:20191108T124500Z
DTEND;VALUE=DATE-TIME:20191108T130000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-38@conference2.aau.at
DESCRIPTION:Speakers: Barbara Pedretscher ()\nModeling fatigue induced deg
radation of a metal film requires a consistent mathematical description of
the physically relevant damage driving forces. In this work\, a state spa
ce model\, based on Itó stochastic processes to account for intrinsic sto
chastic effects\, is followed. Parameter identification and uncertainty qu
antification are based on the system’s corresponding Fokker-Planck equat
ion\, where an adjoint approach as well as profile likelihoods are impleme
nted and compared to MCMC results.\n\nhttps://conference2.aau.at/event/16/
contributions/38/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/38/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Inverse Scheme for Acoustic Source Localization based on Adjoint M
ethod
DTSTART;VALUE=DATE-TIME:20191108T091000Z
DTEND;VALUE=DATE-TIME:20191108T094000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-41@conference2.aau.at
DESCRIPTION:Speakers: Manfred Kaltenbacher ()\nWe propose an inverse schem
e for acoustic source localization based on solving the corresponding part
ial differential equation in the frequency domain (Helmholtz equation) by
applying the Finite Element (FE) method. This allows us to fully take into
account the actual boundary conditions as given in the measurement setup.
To recover the source locations\, an inverse scheme based on a sparsity p
romoting Tikhonov functional to match measured (microphone signals) and si
mulated pressure is proposed. Since the differential operators for the sta
te equation and the adjoint equation are the same\, the FE system matrix f
or both partial differential equations is the same\, which results in a co
mputational highly efficient solution process. Furthermore\, the computati
onal time does not depend on the number of microphones nor on the assumed
number of possible sources. Finally\, the inverse scheme results in a sour
ce map both for amplitude and phase and in addition the reconstructed acou
stic field is provided. The properties of this inverse scheme and its appl
icability to source localization in the low frequency range will be demons
trated.\n\nhttps://conference2.aau.at/event/16/contributions/41/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/41/
END:VEVENT
BEGIN:VEVENT
SUMMARY:On a semismooth* Newton method for generalized equations
DTSTART;VALUE=DATE-TIME:20191107T081000Z
DTEND;VALUE=DATE-TIME:20191107T085000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-42@conference2.aau.at
DESCRIPTION:Speakers: Helmut Gfrerer ()\nIn this talk we introduce a new
notion of semismoothness*\nwhich pertains both sets as well as multifuncti
ons. In the case\nof single-valued maps it is closely related with the sta
ndard\nnotion of semismoothness introduced by Qi and Sun in 1993.\nSemismo
othness* can be equivalently characterized in terms of\nregular\, limiting
and directional limiting coderivatives.\n\nThen we present a semismooth*
Newton method for solving inclusions and\ngeneralized equations (GEs)\, wh
ere the linearization concerns\nboth the single-valued and the multi-value
d part and it is\nperformed on the basis of the respective coderivatives.
Two\nconceptual algorithms will be presented and shown to converge\nlocall
y superlinearly under relatively weak assumptions.\n\nhttps://conference2.
aau.at/event/16/contributions/42/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/42/
END:VEVENT
BEGIN:VEVENT
SUMMARY:On Levenberg-Marquardt type methods for ill-posed operator equatio
ns
DTSTART;VALUE=DATE-TIME:20191107T132000Z
DTEND;VALUE=DATE-TIME:20191107T140000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-43@conference2.aau.at
DESCRIPTION:Speakers: Antonio Leitao ()\nOn this talk we report on a joint
work with Prof. B.Kaltenbacher on Levenberg-Marquardt type methods for so
lving nonlinear ill-posed operator equations in Hilbert spaces.\nMoreover\
, we also discuss some recent research results on this iterative method.\n
\nhttps://conference2.aau.at/event/16/contributions/43/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/43/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Control and controllability of PDEs with hysteresis
DTSTART;VALUE=DATE-TIME:20191107T110000Z
DTEND;VALUE=DATE-TIME:20191107T114000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-44@conference2.aau.at
DESCRIPTION:Speakers: Pavel Krejcí ()\nFor a diffusion equation with a co
mplex hysteresis operator we consider the problem\nof controllability\, th
at is\, finding a control which guarantees that the solution reaches\na de
sired value at a given time. It is solved here by a constructive method ba
sed on\na two-parameter penalty argument. One small parameter penalizes th
e distance of\nthe solution at final time from the expected value\, the se
cond one represents viscous\nregularization of the underlying rate indepen
dent variational inequalities in the hysteresis term.\nWe prove that a sol
ution to the controllability problem can be obtained by passing to the\nsi
ngular limit in the doubly degenerate control system. This is a joint work
\nwith Chiara Gavioli from Modena.\n\nhttps://conference2.aau.at/event/16/
contributions/44/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/44/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Augmented GMRES-type versus CGNE methods for the solution of linea
r ill-posed problems
DTSTART;VALUE=DATE-TIME:20191107T124000Z
DTEND;VALUE=DATE-TIME:20191107T132000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-45@conference2.aau.at
DESCRIPTION:Speakers: Andreas Neubauer ()\nIn this paper we compare (augme
nted) GMRES-type methods and (augmented) CGNE methods. The numerical resul
ts show that the CGNE method is more robust and is suitable for ill-posed
problems with a much higher degree of ill-posedness. GMRES-type methods on
ly yield useful results for very moderate ill-posed problems.\n\nhttps://c
onference2.aau.at/event/16/contributions/45/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/45/
END:VEVENT
BEGIN:VEVENT
SUMMARY:The tangential cone condition for some coefficient identification
model problems in parabolic PDEs.
DTSTART;VALUE=DATE-TIME:20191107T165000Z
DTEND;VALUE=DATE-TIME:20191107T170500Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-46@conference2.aau.at
DESCRIPTION:Speakers: Tram Ngoc Nguyen ()\nThe tangential condition was in
troduced in [Hanke\, Neubauer\, and Scherzer\, 1995] as a sufficient condi
tion for convergence of the Landweber iteration for solving ill-posed prob
lems. This condition ensures nonlinearity of the forward operator fits tog
ether with the data misfit. In this talk\, we present a series of time dep
endent benchmark inverse problems for which we can verify this condition.
We establish the tangential condition in two different modeling settings:
an all-at-once and a reduced version.\n\nhttps://conference2.aau.at/event/
16/contributions/46/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/46/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Tikhonov regularization with oversmoothing penalty: phenomena\, co
nvergence and rates
DTSTART;VALUE=DATE-TIME:20191107T154000Z
DTEND;VALUE=DATE-TIME:20191107T162000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-47@conference2.aau.at
DESCRIPTION:Speakers: Bernd Hofmann ()\nThe presentation is devoted to Tik
honov regularization under conditional stability\nestimates for nonlinear
ill-posed operator equations in Hilbert scales. Our focus\nis on the case
of oversmoothing penalties\, for which the true solution no longer\nattain
s a finite value. In this context\, we present some new results on converg
ence\nand recall assertions on rates. We strongly highlight the local char
acter of\nconditional stability estimates for nonlinear problems and demon
strate that\npitfalls may occur. Then convergence can completely fail and
the stabilizing\neffect of conditional stability may be lost. Numerical ca
se studies for some\nnonlinear examples illustrate such effects.\n\nThis t
alk presents joint work with Peter Mathé (Berlin)\, Robert Plato (Siegen)
\,\nDaniel Gerth and Christopher Hofmann (Chemnitz). Research is supported
by the\nDeutsche Forschungsgemeinschaft (DFG) under grant HO 1454/12-1.\n
\nhttps://conference2.aau.at/event/16/contributions/47/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/47/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A Collection of Inverse Problems in Engineering Applications
DTSTART;VALUE=DATE-TIME:20191108T103000Z
DTEND;VALUE=DATE-TIME:20191108T110000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-48@conference2.aau.at
DESCRIPTION:Speakers: Tom Lahmer ()\nThe talk presents a series of example
s\, where engineers are actually faced with inverse problems of different
types. Solution strategies and results are given for several examples incl
uding poroelasticity and piezoelectricity. The inverse problems are mainly
of parameter and source identification type.\n\nA new series of inverse p
roblems comes up by the rapid developments in additive printing techniques
. Here\, the talk will name some of the newly arising challenges during th
e characterization of materials which are built by additive printing techn
iques.\n\nhttps://conference2.aau.at/event/16/contributions/48/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/48/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A Random Wanderers guide to the Diffusion Equation Universe
DTSTART;VALUE=DATE-TIME:20191108T080000Z
DTEND;VALUE=DATE-TIME:20191108T084000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-49@conference2.aau.at
DESCRIPTION:Speakers: Bill Rundell ()\nThe classical reaction-diffusion eq
uation can be stated as the rate of change of the state variable $u$ equal
s the sum of a diffusion operator and a (often nonlinear) reaction term: $
u_t = -L\\\,u + f(u)$\, where typically $-L$ is an elliptic operator. The
equation\, while known in form\, can have specific parameters undetermined
\; for example coefficients in $L$ or the function $f$ itself.\nWe look at
various models for recovery of these but we also do so in a context beyon
d Brownian motion diffusion and its parabolic pde format. These so-called
anomalous diffusion models give rise to nonlocal differential operators of
fractional type and add further difficulties to an already complex proble
m.\nThe work is joint with Barbara Kaltenbacher.\n\nhttps://conference2.aa
u.at/event/16/contributions/49/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/49/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A multiplicative iterative method for ill-posed problems
DTSTART;VALUE=DATE-TIME:20191107T162000Z
DTEND;VALUE=DATE-TIME:20191107T165000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-51@conference2.aau.at
DESCRIPTION:Speakers: Elena Resmerita ()\nWe present a multiplicative entr
opic type method for ill-posed equations\, which has a closed form and pre
serves nonnegativity of the iterates. Interestingly\, this has been invest
igated in finite dimensional optimization under different names/versions\,
while related methods have been considered in inverse problems as well. W
e discuss convergence and error estimates\, and test the method on several
examples\, in comparison with two other iterative methods.\n\nhttps://co
nference2.aau.at/event/16/contributions/51/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/51/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Data driven regularization by projection
DTSTART;VALUE=DATE-TIME:20191108T084000Z
DTEND;VALUE=DATE-TIME:20191108T091000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-50@conference2.aau.at
DESCRIPTION:Speakers: Otmar Scherzer ()\nRegularization methods for invers
e problems can be based on mathematical forward methods which represent th
e Physics and Chemistry\nas precise as possible. With the rise of the area
of big data\, methods that combine forward modelling with data driven tec
hniques have been\nbeing developed. In this talk we demonstrate that regul
arisation by projection can be formulated in a purely data driven setting\
nwhen the linear forward operator is given only through training data. We
study convergence and stability of the regularised solutions.\nWe discuss
counter examples on convergence of the method of regularization by project
ion by Seidman in this context.\n\nThis is joint work with Andrea Aspri (R
ICAM\, Linz)\, Yury Korolev (Cambridge).\n\nhttps://conference2.aau.at/eve
nt/16/contributions/50/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/50/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mathematical modelling and statistical inference for large-scale m
odels of cancer signalling
DTSTART;VALUE=DATE-TIME:20191108T100000Z
DTEND;VALUE=DATE-TIME:20191108T103000Z
DTSTAMP;VALUE=DATE-TIME:20200814T070812Z
UID:indico-contribution-16-52@conference2.aau.at
DESCRIPTION:Speakers: Jan Hasenauer ()\nThe sequencing of cancer biopsies
revealed that cancer is multi-factorial diseases\, which strongly vary bet
ween patients. This inter-patient variability poses a challenge for clinic
ians. A priori it is not clear which drug will be most beneficial for a sp
ecific patient. Here\, we approach the problem of drug response prediction
using mechanistic mathematical models. We develop a mathematical model de
scribing several cancer associated signaling pathways. This model can be i
ndividualized using sequencing data. For statistical inference we develop
a scalable approach facilitating the study of models with thousands of par
ameters. We the approach to analyse drug response data from the Cancer Cel
l Line Encyclopaedia for 7 drugs and 120 cell lines originating from five
different tissues. These results demonstrate the potential of large-scale
mechanistic modeling for drug selection in personalized therapy.\n\nhttps:
//conference2.aau.at/event/16/contributions/52/
LOCATION:Stiftungssaal K.0.01
URL:https://conference2.aau.at/event/16/contributions/52/
END:VEVENT
END:VCALENDAR