51社区黑料

2014-2015 Operations Research Seminar

Welcome to the Web page of the 51社区黑料Operations Research Seminar Series. We are associated with:
, and
.
Our aim is to meet and discuss Operations Research topics.

Unless noted the talks will be at 2:30 on Thursday in Room 5380 (in Spring), 51社区黑料Surrey.
Please contact if you would like to speak.

Date Speaker Title and Abstract

Apr. 16th

via COCANA (Kelowna)

and

Nonconvex Bundle Method for Constrained Optimization Problems

Further details available from the Website.

Apr. 10th

*10:30*

*SUR 3250* M. Beddis, M. Mitrovic and M. Sharma

Math 402W Operations Research Clinic project presentation

Selecting Station Locations for a Public Bike-Share Program: A Case Study for the City of Vancouver, B.C.

Apr. 9th

Geometric versions of the 3-dimensional assignment problem

Abstract:
In this talk we will discuss the computational complexity of special cases of the 3-dimensional assignment problem where the elements are points in a Cartesian space and where the cost coefficients are the perimeters of the corresponding triangles measured according to a certain norm. All our results also carry over to the corresponding special cases of the 3-dimensional matching problem. The minimization version is NP-hard for every norm, even if the underlying Cartesian space is 2-dimensional. The maximization version is polynomially solvable, if the dimension of the Cartesian space is fixed and if the considered norm has a polyhedral unit ball. If the dimension of the Cartesian space is part of the input, the maximization version is NP-hard for every L_p norm.
This is joint work with Bettina Klinz and Gerhard Woeginger, and a preprint is available .

Apr. 2nd

via COCANA (Kelowna)

Jim Nastos

Observations on problem reductions

Further details available from the Website.

Mar. 26th

Bounds on Eigenvalues of Matrices Arising from Interior-Point Methods

Abstract:
Interior-point methods feature prominently among numerical methods for inequality-constrained optimization problems, and involve the need to solve a sequence of linear systems that typically become increasingly ill-conditioned with the iterations. To solve these systems, whose original form has a nonsymmetric 3-by-3 block structure, it is common practice to perform block elimination and either solve the resulting reduced saddle-point system, or further reduce the system to the normal equations and apply a symmetric positive definite solver. In this talk we use energy estimates to obtain bounds on the eigenvalues of the matrices, and conclude that the original unreduced matrix has more favorable eigenvalue bounds than the alternative reduced versions. Our analysis includes regularized variants of those matrices that do not require typical regularity assumptions. This is joint work with Erin Moulding and Dominique Orban.

Mar. 19th

via COCANA (Kelowna)

TBA

Further details available from the Website.

Mar. 5th

via COCANA (Kelowna)

Tractable Big Data and Big Models in Machine Learning

Further details available from the Website.

Feb. 26th

via COCANA (Kelowna)

Walaa Moursi

On the range of the Douglas-Rachford operator

Further details available from the Website.

Feb. 19th

via COCANA (Kelowna)

Improved Density Estimation via Data Sharpening

Further details available from the Website.

Feb. 5th

On finite convergence of an explicit exchange method for convex semi-infinite programming problems with second-order cone constraints

Abstract:
In this talk, we propose an explicit exchange algorithm for solving semi-infinite programming problem (SIP) with second-order cone (SOC) constraints. We prove, by using the slackness complementarity conditions, that the algorithm terminates in a finite number of iterations and the obtained solution sufficiently approximates the original SIP solution. In existing studies on SIPs, only the nonnegative constraints were considered, and hence, the slackness complementarity conditions were separable to each component. However, in our study, the existing componentwise analyses are not applicable anymore since the slackness complementarity conditions are associated with SOCs. In order to overcome such a difficulty, we introduce a certain coordinate system based on the spectral factorization. In the numerical experiments, we solve some test problems to see the effectiveness of the proposed algorithm.

Jan. 22nd

via COCANA (Kelowna)

Chayne Planiden

Moreau Envelopes and Thresholds of Prox-boundedness

Further details available from the Website.

Jan. 8th

*SUR 5380*

Salvo Models for Missile Combat

Abstract:
Modern surface warships attack and defend using guided missiles such as the Harpoon and Standard. Because few battles have been fought this way, missile combat is not as well understood as that involving gunfire. Salvo models provide a simple way to represent such battles, much as Lanchester models represent gunfire battles. This talk will introduce salvo combat models, describe some of their properties, and demonstrate their application to the carrier airstrikes of the 1942 Battle of the Coral Sea.

Dec. 4th

via COCANA (Kelowna)

Applications of Modeling and Optimization Tools for Quality Improvement in Composites Manufacturing

Further details available from the Website.

Tuesday,
Nov. 25th

Joint O.R. and Discrete Math Seminar

*Burnaby*
*AQ K9509*

Flat Norm Decomposition of Integral Currents

Abstract:
Currents represent generalized surfaces studied in geometric measure theory. The flat norm provides a natural distance in the space of currents, and works by decomposing a d-dimensional current into d- and (the boundary of) (d+1)-dimensional pieces. A natural question about currents is the following. If the input is an integral current, i.e., a current with integer multiplicities, can its flat norm decomposition be integral as well? The answer is not known in general, except in the case of d-currents that are boundaries of (d+1)-currents in (d+1)-dimensional space. On the other hand, for the discretization of the flat norm on a finite simplicial complex, the analogous statement remains true even when the inputs are not boundaries. This result is implied by the boundary matrix of the simplicial complex being totally unimodular, guaranteeing integer solutions for an associated integer linear program. We develop an analysis framework that extends the result in the simplicial setting to that for d-currents in (d+1)-dimensional space, provided a suitable triangulation result holds. Following results of Shewchuk on triangulating planar straight line graphs, our framework shows that the discrete result implies the continuous result for the case of 1-currents in 2D space.

This is joint work with Sharif Ibrahim and Kevin Vixie, and a preprint is available .

Tuesday,
Nov. 25th

*10:00 a.m.*

*SUR 5060* Piyashat Sripratak

The Bipartite Boolean Quadratic Programming Problem

Ph.D. thesis defence

Nov. 20th

via COCANA (Kelowna)

Wasserstein barycenters and related problems: theory, numerics and applications

Further details available from the Website.

Oct. 16th

via COCANA (Kelowna)

Lying with Statistics! Optimally Changing Data to Improve Nonparametric Function Estimates

Further details available from the Website.

Oct. 2nd

via COCANA (Kelowna)

On a shortest 2-path problem

Further details available from the Website.

Sept. 21st

*8:30-4:30*

Hosted by

Details of the Fall 2014 West Coast Optimization Meeting .

Sept. 11th

via COCANA (Kelowna)

Smoothing SQP methods for solving nonsmooth and nonconvex constrained optimization problems

Further details available from the Website.

Aug. 26th

*10:00-12:00*

*SUR 5380*

Ph.D. thesis defence

Senior Supervisor: Z.Lu Optimization Methods for Sparse Approximation

Archives of the 2006-07, 2007-08, 2008-09, 2009-10 2010-11 2011-12 and 2013-14 51社区黑料Operations Research Seminars.

Last modified September 6th, 2014.

Date	Speaker	Title and Abstract
Apr. 16th via COCANA (Kelowna)	and	Nonconvex Bundle Method for Constrained Optimization Problems Further details available from the Website.
Apr. 10th *10:30* *SUR 3250*	M. Beddis, M. Mitrovic and M. Sharma	Math 402W Operations Research Clinic project presentation Selecting Station Locations for a Public Bike-Share Program: A Case Study for the City of Vancouver, B.C.
Apr. 9th		Geometric versions of the 3-dimensional assignment problem Abstract: In this talk we will discuss the computational complexity of special cases of the 3-dimensional assignment problem where the elements are points in a Cartesian space and where the cost coefficients are the perimeters of the corresponding triangles measured according to a certain norm. All our results also carry over to the corresponding special cases of the 3-dimensional matching problem. The minimization version is NP-hard for every norm, even if the underlying Cartesian space is 2-dimensional. The maximization version is polynomially solvable, if the dimension of the Cartesian space is fixed and if the considered norm has a polyhedral unit ball. If the dimension of the Cartesian space is part of the input, the maximization version is NP-hard for every L_p norm. This is joint work with Bettina Klinz and Gerhard Woeginger, and a preprint is available .
Apr. 2nd via COCANA (Kelowna)	Jim Nastos	Observations on problem reductions Further details available from the Website.
Mar. 26th		Bounds on Eigenvalues of Matrices Arising from Interior-Point Methods Abstract: Interior-point methods feature prominently among numerical methods for inequality-constrained optimization problems, and involve the need to solve a sequence of linear systems that typically become increasingly ill-conditioned with the iterations. To solve these systems, whose original form has a nonsymmetric 3-by-3 block structure, it is common practice to perform block elimination and either solve the resulting reduced saddle-point system, or further reduce the system to the normal equations and apply a symmetric positive definite solver. In this talk we use energy estimates to obtain bounds on the eigenvalues of the matrices, and conclude that the original unreduced matrix has more favorable eigenvalue bounds than the alternative reduced versions. Our analysis includes regularized variants of those matrices that do not require typical regularity assumptions. This is joint work with Erin Moulding and Dominique Orban.
Mar. 19th via COCANA (Kelowna)		TBA Further details available from the Website.
Mar. 5th via COCANA (Kelowna)		Tractable Big Data and Big Models in Machine Learning Further details available from the Website.
Feb. 26th via COCANA (Kelowna)	Walaa Moursi	On the range of the Douglas-Rachford operator Further details available from the Website.
Feb. 19th via COCANA (Kelowna)		Improved Density Estimation via Data Sharpening Further details available from the Website.
Feb. 5th		On finite convergence of an explicit exchange method for convex semi-infinite programming problems with second-order cone constraints Abstract: In this talk, we propose an explicit exchange algorithm for solving semi-infinite programming problem (SIP) with second-order cone (SOC) constraints. We prove, by using the slackness complementarity conditions, that the algorithm terminates in a finite number of iterations and the obtained solution sufficiently approximates the original SIP solution. In existing studies on SIPs, only the nonnegative constraints were considered, and hence, the slackness complementarity conditions were separable to each component. However, in our study, the existing componentwise analyses are not applicable anymore since the slackness complementarity conditions are associated with SOCs. In order to overcome such a difficulty, we introduce a certain coordinate system based on the spectral factorization. In the numerical experiments, we solve some test problems to see the effectiveness of the proposed algorithm.
Jan. 22nd via COCANA (Kelowna)	Chayne Planiden	Moreau Envelopes and Thresholds of Prox-boundedness Further details available from the Website.
Jan. 8th *SUR 5380*		Salvo Models for Missile Combat Abstract: Modern surface warships attack and defend using guided missiles such as the Harpoon and Standard. Because few battles have been fought this way, missile combat is not as well understood as that involving gunfire. Salvo models provide a simple way to represent such battles, much as Lanchester models represent gunfire battles. This talk will introduce salvo combat models, describe some of their properties, and demonstrate their application to the carrier airstrikes of the 1942 Battle of the Coral Sea.
Dec. 4th via COCANA (Kelowna)		Applications of Modeling and Optimization Tools for Quality Improvement in Composites Manufacturing Further details available from the Website.
Tuesday, Nov. 25th Joint O.R. and Discrete Math Seminar Burnaby AQ K9509		Flat Norm Decomposition of Integral Currents Abstract: Currents represent generalized surfaces studied in geometric measure theory. The flat norm provides a natural distance in the space of currents, and works by decomposing a d-dimensional current into d- and (the boundary of) (d+1)-dimensional pieces. A natural question about currents is the following. If the input is an integral current, i.e., a current with integer multiplicities, can its flat norm decomposition be integral as well? The answer is not known in general, except in the case of d-currents that are boundaries of (d+1)-currents in (d+1)-dimensional space. On the other hand, for the discretization of the flat norm on a finite simplicial complex, the analogous statement remains true even when the inputs are not boundaries. This result is implied by the boundary matrix of the simplicial complex being totally unimodular, guaranteeing integer solutions for an associated integer linear program. We develop an analysis framework that extends the result in the simplicial setting to that for d-currents in (d+1)-dimensional space, provided a suitable triangulation result holds. Following results of Shewchuk on triangulating planar straight line graphs, our framework shows that the discrete result implies the continuous result for the case of 1-currents in 2D space. This is joint work with Sharif Ibrahim and Kevin Vixie, and a preprint is available .
Tuesday, Nov. 25th 10:00 a.m. SUR 5060	Piyashat Sripratak	The Bipartite Boolean Quadratic Programming Problem Ph.D. thesis defence
Nov. 20th via COCANA (Kelowna)		Wasserstein barycenters and related problems: theory, numerics and applications Further details available from the Website.
Oct. 16th via COCANA (Kelowna)		Lying with Statistics! Optimally Changing Data to Improve Nonparametric Function Estimates Further details available from the Website.
Oct. 2nd via COCANA (Kelowna)		On a shortest 2-path problem Further details available from the Website.
Sept. 21st 8:30-4:30	Hosted by	Details of the Fall 2014 West Coast Optimization Meeting .
Sept. 11th via COCANA (Kelowna)		Smoothing SQP methods for solving nonsmooth and nonconvex constrained optimization problems Further details available from the Website.
Aug. 26th 10:00-12:00 *SUR 5380*	Ph.D. thesis defence Senior Supervisor: Z.Lu	Optimization Methods for Sparse Approximation