NA Digest, V. 20, # 4

NA Digest Sunday, January 26, 2020 Volume 20 : Issue 4


Today's Editor:

Daniel M. Dunlavy
Sandia National Labs
[email protected]

Today's Topics: Subscribe, unsubscribe, change address, or for na-digest archives: http://www.netlib.org/na-digest-html/faq.html

Submissions for NA Digest:

http://icl.utk.edu/na-digest/


From: Philippe Toint [email protected]
Date: January 20, 2020
Subject: BFO (Brute Force Optimizer) Release 2.0 available


Margherita Porcelli and Philippe Toint have the pleasure to inform you
that Release 2.0 of the BFO (Brute Force Optimizer) Matlab package is
now available...and that it merits its name a little bit less.

BFO's remarkable features are:
- its ability to handle a mix of continuous, discrete and categorical
variables,
- a novel self-training option, which may significantly improve the
performance of the code on the user's class of problems, its
capacity to handle multilevel, min-max and equilibrium problems,
- an extremely versatile and easy-to-use interface. BFO also provides
a number of user-oriented features, such as checkpointing and
restart, facilities for specifying variable scaling, parameters
passing, etc.
- tools allowing the user to specify a variety of termination
conditions,
- BFOSS, a library whose purpose is to compute interpolation-based
search steps.
- using coordinate partially-separable problem structure. This
ubiquitous problem structure can now be exploited by BFO, leading to
very significant gains in performance (orders of magnitude) also
allowing the use of BFO for large problems (thousands of variables
and more).

Release 2.0 of the Matlab BFO package is a major upgrade from Release
1 (https://sites.google.com/site/bfocode/home]) and includes several
important new problem-oriented features. It also improves performance
and stability upon Release 1.0 and corrects a few bugs.

The code is publicly available on GITHUB
(https://github.com/m01marpor/BFO).

We welcome comments and suggestions.



From: GEORGE ANASTASSIOU [email protected]
Date: January 22, 2020
Subject: New Book, Intelligent Analysis


George A. Anastassiou, Intelligent Analysis: Fractional Inequalities
and Approximations Expanded, SPRINGER NATURE, 2020

Computational and fractional analysis play a very important role in
now days either by themselves or because they cover a great variety of
applications in the real world. We start with the important Iyengar
type inequalities and we continue with Choquet integral analytical
inequalities, related to great applications in economics. We deal with
the local fractional derivatives of Riemann-Liouville type and related
results including inequalities. We encounter the case of low order
Riemann-Liouville fractional derivatives and inequalities without
initial conditions, we study also related approximations. We continue
with the quantitative complex approximation theory by operators and
various important complex fractional inequalities. We also deal with
the conformable fractional approximation of well-known Csiszar's
f-divergence. We present conformable fractional self adjoint operator
inequalities. We continue with the study of new local fractional
M-derivatives which carry all the basic properties of ordinary
derivatives. We finish with the new complex multivariate Taylor
formula with integral remainder. This book results are expected to
find applications in many areas of pure and applied mathematics. As
such this monograph is suitable for researchers, graduate students,
and seminars of the above disciplines, also to be in all science and
engineering libraries.


From: Gerhard-Wilhelm Weber [email protected]
Date: January 20, 2020
Subject: Deadline Extended, IFORS 2020, South Korea, Jun 2020


2020 IFORS Conference: abstract submissions now open until 31st
January 2020

The Organizing and Program Committees for IFORS 2020 welcomes all
operations researchers to attend the 22nd Conference of the
International Federation of Operational Research Societies (IFORS).

Dates: June 21-26 (Sun-Fri), 2020
Location: Seoul, Korea
Submit an abstract: https://www.euro-online.org/conf/ifors2020/

Please visit the IFORS 2020 webpage (www.ifors2020.kr) to learn more
about important dates, abstract submission, registration, special
issues of journals, the social program and accommodation.

The IFORS 2020 conference will highlight global developments in
operations research and show how the tools of operations research are
expanding their impact on society, health, science and industry. IFORS
conferences, held only once every three years, provide a platform for
experts from around the world to showcase the diverse potential of
state-of-the-art operations research techniques and technologies.
IFORS 2020 offers a unique opportunity to network with operations
research analysts, industrial users of operations research, and
academic and industry experts from all parts of the globe.



From: Nick Higham [email protected]
Date: January 20, 2020
Subject: Workshop Celebrating 70th Birthday of Jack Dongarra, UK, Jul 2020


New Directions in Numerical Linear Algebra and High Performance
Computing: Celebrating the 70th Birthday of Jack Dongarra, July 17,
2020, University of Manchester.

This international workshop focuses on numerical linear algebra and
high performance computing and brings together researchers working in
these areas to discuss current developments and challenges in the
light of evolving computer hardware. The workshop is being held to
honor Jack Dongarra on the occasion of his 70th birthday.

Contributed posters are welcome and for full consideration should be
submitted by May 15, 2020, as part of the registration process.

For details of the conference, including the invited speakers, and
registration (deadline June 30, 2020) see
https://nla-group.org/new-directions-in-numerical-linear-algebra-and-high-performance-
computing-2020/


From: Frédéric JEAN [email protected]
Date: January 20, 2020
Subject: Faculty Position, Optimization, ENSTA Paris


ENSTA Paris opens a faculty position in optimization, see details here
(in French):

https://uma.ensta-paris.fr/var/uma/trainings/2020/prof_optim_uma.pdf

The position a priori requires a fairly good level of French, but
excellent non-French speaking candidates will be considered.

The deadline for applications is Friday, March 15, 2020.


From: John Stockie [email protected]
Date: January 22, 2020
Subject: Postdoc Position, Mathematics, Simon Fraser Univ


Alan Mekler Postdoctoral Fellowship, Mathematics Department
Simon Fraser University

Applications are invited for the Alan Mekler Postdoctoral Fellowship
in the Department of Mathematics at Simon Fraser University. This
fellowship provides the opportunity to engage in mathematical research
and undergraduate teaching, and is targeted towards talented
mathematicians who have recently completed their PhD (in 2017 or
later). The initial appointment will be for one year, renewable for a
second year subject to satisfactory performance in both research and
teaching. The fellowship is open to candidates of any nationality and
the selection decision will be based upon research potential, teaching
ability, and the fit with current areas of active research in the
department. A list of Mathematics faculty and their research interests
can be found here: http://www.sfu.ca/math/research

SFU is dedicated to fostering a culture of inclusion and mutual
respect, and to advancing diversity in the workplace by encouraging
applications from all qualified individuals including women, persons
with disabilities, visible minorities, Indigenous Peoples, and sexual
and gender diverse communities. The expected start date for the
fellowship is 1 September 2020 and the annual stipend is a minimum of
$55,000, plus a $3,000 grant for research expenses. The stipend
includes a commitment to teach two undergraduate courses per year. Any
questions about the position or application procedure can be addressed
to Tom Archibald at [email protected].

How To Apply: Applications should be submitted electronically through
MathJobs at http://www.mathjobs.org/jobs/jobs/15569, and must consist
of: Cover letter, which indicates the names of up to three SFU faculty
members whose research is closest to the applicant's; Curriculum
vitae, including a list of publications; Research statement; Teaching
statement; Three letters of reference submitted directly by the referees.
At least one letter must report on the applicant's teaching ability.

Completed applications will be reviewed on an ongoing basis starting
29 February 2020, and will be accepted until the position is
filled. If necessary, applications may be sent by postal mail to the
address below but these applications should arrive by the same
deadline:

Alan Mekler Postdoctoral Fellow Search
c/o Prof. T. Archibald, Chair
Department of Mathematics
Simon Fraser University
8888 University Drive
Burnaby, BC, Canada, V5A 1S6


From: Tuomo Valkonen [email protected]
Date: January 24, 2020
Subject: Postdoc Position, Nonsmooth Optimization/Inverse Problems, Univ of Helsinki


I am looking for a post-doctoral researcher to work on nonsmooth
optimisation methods specifically for the solution of inverse
problems. The position is at the Department of Mathematics and
Statistics at the University of Helsinki in close contact with the
Finnish Centre of Excellence in Inverse Modelling and Imaging. The
starting date is negotiable, preferably in September 2020. In the
first place, funding is secured until the end of August 2021. An
extension is foreseen.

Enquiries: Tuomo Valkonen

To apply:
https://www.helsinki.fi/en/open-positions/postdoctoral-researcher-in-optimization-for-
inverse-problems


From: Jack S. Hale [email protected]
Date: January 23, 2020
Subject: Postdoc Position, Viscoelastic Flow Modelling, Univ of Luxembourg


A postdoctoral position in data-enhanced models for simulating
industrial extrusion processes is available immediately.

This research project aims to improve the accuracy of swell prediction
in industrial extrusion processes using a combined physics and
data-driven modelling approach.

See http://emea3.mrted.ly/2e627 for more details.



From: Nail Yamaleev [email protected]
Date: January 22, 2020
Subject: PhD Position, Computational fluid dynamics, Old Dominion Univ


Applications are invited for a PhD student position in the Department
of Mathematics and Statistics at Old Dominion University (Norfolk,
VA). This position will provide a unique opportunity to work on a
cutting-edge project in the group of Prof. N. Yamaleev in close
collaboration with research scientists of NASA Langley Research
Center. Current research in the group focuses on the development of
new entropy stable schemes for the Navier-Stokes equations,
adjoint-based methods for PDE-constrained optimization problems, and
grid adaptation methods based on error minimization. Our group has a
history of producing highly educated, independent, exceptionally
talented PhD scientists and postdocs. More information about our
research can be found at:
https://www.odu.edu/directory/people/n/nyamalee#profiletab=1
A typical PhD study in our group leads to participation in national
and international conferences and meetings, multiple publications in
top journals such as Journal of Computational Physics, Computers &
Fluids, AIAA Journal etc., and ample opportunities for networking with
leading research scientists from NASA, national labs, academia, and
industry.

We are looking for an enthusiastic and highly motivated PhD candidate
with a M.S. or B.S. degree in Mathematics, Computer Science,
Engineering or a closely related field. A solid background in
numerical methods, excellent programming skills (Fortran 90 or C++),
and effective communication skills (written/spoken English) are
required. Interested candidates should apply for a graduate
assistantship in computational and applied mathematics at:
https://www.odu.edu/admission/graduate Further details on how to apply
can be found at:
http://catalog.odu.edu/graduate/collegeofsciences/mathematicsstatistics
/#doctorofphilosophy-computationalandappliedmathematics For more
information, please contact Dr. Yamaleev at [email protected].



From: Gabriel Barrenechea [email protected]
Date: January 24, 2020
Subject: PhD Position, Finite element methods for coupled problems, Strathclyde


Jennifer Pestana and I are currently advertising a fully funded
3.5-year PhD studentship on the topic "Finite element methods for
coupled problems". The project lies in the interface between deriving
new discretisations and proposing efficient linear (and nonlinear)
solvers for the resulting large algebraic systems. As such, a
candidate with a good knowledge of at least one of these two topics
would be ideal.

The position is open to any UK/EU student, and it can be filled as
soon as an appropriate candidate is selected. The student is expected
to start around the 01/04/2020.

If you know of anyone that may be interested (and would be
appropriate), please let her/him know. We require the student's CV,
two reference letters, and a cover letter. We will also interview the
candidates before making a decision.

Please send applications and informal enquiries to
[email protected]


From: Eskil Hansen [email protected]
Date: January 24, 2020
Subject: PhD Position, Numerical Analysis, Lund Univ


The Centre for Mathematical Sciences, Lund University, is recruiting a
PhD student in numerical analysis for the project ''Next generation
numerical partitioning schemes for time dependent PDEs''.

Temporary position for 5 years
Application deadline: February 12, 2020
Earliest starting date: March 15, 2020

For more information visit
https://lu.varbi.com/en/what:job/jobID:310970/


From: Heike Sill [email protected]
Date: January 23, 2020
Subject: PhD Position, WIAS, Berlin


The Research Group "Partial Differential Equations" (Head:
Prof. Dr. A. Mielke) at WIAS offers a PhD Student Position (f/m/d)
(Ref. 20/03). to be filled as soon as possible. The position is
associated with the research project "Pattern formation in coupled
parabolic systems" within the DFG Collaborative Research Center SFB910
"Control of self-organizing nonlinear systems: Theoretical methods and
concepts of applications". The prerequisites are a diploma or master
degree in mathematics. The successful candidate will have sound
knowledge in one of the following fields: functional analysis,
nonlinear partial differential equations, reaction-diffusion systems,
mathematical modeling in fluid or solid mechanics. Scientific
question may be addressed to Prof. A. Mielke
([email protected]). The appointment is limited to 2
years and may be extended. The salary is according to the German TVoeD
scale with a working time of 29.25 hours per week. The Institute aims
to increase the proportion of women in this field, so applications
from women are particularly welcome. Among equally qualified
applicants, disabled candidates will be given preference.

Please upload your complete application documents (motivation letter,
detailed CV, certificates, list of MSc courses and grades, copy of the
master thesis, reference letter, etc.) via our website or via the
applicant portal until February 29th, 2020 using the button "Apply
online".

See here for more information: https://short.sg/j/5908710


From: Andreas Adelmann [email protected]
Date: January 22, 2020
Subject: PhD Positions, Computational Physics, Paul Scherrer Institut/ETH Zurich


1. COmbined Loading Optimization with Simulations and Surrogate models
(COLOSS). The COLOSS project aims at an integral computational scheme
for in- core/out-of-core spent fuel analysis, i.e. the development of
so-called multidimensional loading surfaces (MLS). Considering the 5
nuclear power plants in Switzerland, about 12 000 spent fuel
assemblies are expected for the final repository. In each canister for
PWRs there will be 4 assemblies, hence the number of possible
arrangements is over 10E21 possibilities. On this enormous amount of
configurations boundary conditions w.r.t. safety i.e. effective
criticality and maximum allowed decay heat, must be applied. On top
of this the time dependent nature of the decay needs to be taken into
account. In order to achieve a best-estimate-plus-uncertainty (BEPU)
methodology for such MLS of storage/transport casks, the COLOSS
project will aim at deploying data science techniques as alternative
to the computationally unaffordable forward monte-carlo based
modelling and uncertainty propagation methods. This PhD thesis project
will be an integral part of COLOSS. The aim is to develop surrogate
models using the reference numerical simulations as basis and
thereafter, apply the surrogate models for MLS uncertainty/
sensitivity analyses and/or as loading optimisation method. This will
contribute to an optimal experimental design (numerical simulations),
constraints on design space (type and number of dependant variables)
and accuracy versus computational performance of the employed methods.
Your ideally have a sound background in theoretical or reactor
physics, statistical & machine learning methods, numerical mathematics
and be fluent in Python programming. For more information please
contact Andreas Adelmann ([email protected]) or Dimitri
A. Rochman ([email protected])

2. Fast AI driven plan generation for patient selection in proton
therapy. Radiotherapy can be considered the epitome of the concept of
'personalised medicine'. Using advanced computer methods based on
physical models of interactions of radiation with tissue, a customised
treatment is designed a priori for every patient presenting for
radiotherapy. Proton therapy facilities, still are a rare resource,
with the existing facilities having limited capacity for treating
large numbers of patients. As such, patient selection is of critical
importance if the benefits of proton therapy are to be best exploited.
This research project is about an efficient and effective approach for
planning high quality and clinically realistic treatment plans, with a
minimal need for human interaction, and with turn-around times in the
order of minutes. A conventional computational approach, will in the
foreseen future, not be able to meet the stringent requirements
concerning "time to solution" i.e. minutes. In this research project
we will study statistical and machine learning approaches towards
automatic planning systems for patient selection. The Center for
Proton Therapy has the longest experience of clinical pencil beam
scanning proton therapy, and hence has one of the largest databases of
clinically validated patient treatments in the world. It is therefore
in an ideal position to provide high quality data for a machine
leaning approach to comprehensive and clinically relevant automatic
treatment planning and patient selection. We are searching for a
highly motivated student working at the intersection of machine &
statistical learning and medical-physics. The ideal candidate has a
great affinity to mathematical modelling with a sound background in
machine and statistical learning methods. A sound background in either
physics or medical-physics is needed. For more information please
contact Andreas Adelmann ([email protected]) or Tony Lomax
([email protected])


From: Erin Carson [email protected]
Date: January 21, 2020
Subject: PhD Positions, Numerical Linear Algebra and HPC, Charles Univ


Funded PhD positions are available within the framework of the Primus
Research Program "Scalable and Accurate Numerical Linear Algebra for
Next-Generation Hardware", led by Dr. Erin Carson at the Faculty of
Mathematics and Physics at Charles University. Further details about
the project can be found at the project website:
https://www.karlin.mff.cuni.cz/~carson/primus.html.

Applications are invited from candidates who have strong background in
numerical linear algebra, numerical analysis, parallel computing, or
computational/data science application domains. Successful candidates
will formally enroll in the PhD program at Charles University.

Preliminary applications are due March 22, 2020. For details on how to
apply, see https://www.mathjobs.org/jobs/jobs/15563


From: Claude Brezinski [email protected]
Date: January 23, 2020
Subject: Contents, Numerical Algorithms, 83 (1)


Table of Contents
Numerical Algorithms, Vol. 83, No. 1

Differentiation matrices for univariate polynomials, Amirhossein
Amiraslani, Robert M. Corless, Madhusoodan Gunasingam

Fast discrete convolution in R^2 with radial kernels using non-uniform
fast Fourier transform with nonequispaced frequencies, Martin Averseng

Decomposition into subspaces preconditioning: abstract framework,
Jakub Hrncir, Ivana Pultarova, Zdenek Strakos

A relaxation-type Galerkin FEM for nonlinear fractional Schrodinger
equations, Meng Li, Chengming Huang, Wanyuan Ming

A priori error estimates of expanded mixed FEM for Kirchhoff type
parabolic equation, Nisha Sharma, Morrakot Khebchareon, Amiya K. Pani

A breakdown-free algorithm for computing the determinants of periodic
tridiagonal matrices, Ji-Teng Jia

A quadratic spline collocation method for the Dirichlet biharmonic
problem, Bernard Bialecki, Graeme Fairweather, Andreas Karageorghis,
Jonathan Maack

Modulus-based matrix splitting methods for horizontal linear
complementarity problems, Francesco Mezzadri, Emanuele Galligani

Computing the Lambert W function in arbitrary-precision complex
interval arithmetic, Fredrik Johansson

Chebyshev spectral collocation method for system of nonlinear Volterra
integral equations, Zhendong Gu

A study of Schroeder's method for the matrix pth root using power
series expansions, Chun-Hua Guo, Di Lu

Accelerated double-step scale splitting iteration method for solving a
class of complex symmetric linear systems, Mehdi Dehghan, Akbar
Shirilord

New self-adaptive step size algorithms for solving split variational
inclusion problems and its applications, Yan Tang, Aviv Gibali

Extending the applicability of the inexact Newton-HSS method for
solving large systems of nonlinear equations, Ioannis K. Argyros,
Santhosh George, Kedarnath Senapati

Note on error bounds for linear complementarity problems of Nekrasov
matrices, Chaoqian Li, Shaorong Yang, Hui Huang, Yaotang Li, Yimin Wei

A structural analysis of field/circuit coupled problems based on a
generalised circuit element, Idoia Cortes Garcia, Herbert De Gersem,
Sebastian Schoeps

GPU acceleration of splitting schemes applied to differential matrix
equations, Hermann Mena, Lena-Maria Pfurtscheller, Tony Stillfjord

End of Digest
**************************