Date  Speaker  Title and Abstract 
Tuesday,
June 4th Joint O.R. and Discrete Math Seminar *Burnaby* *1:30* *AQ K9509* 
Ahmad Abdi
Tepper School of Business Carnegie Mellon University and Department of Mathematics London School of Economics 
Ideal clutters and kwise intersecting families
Abstract: A clutter is *ideal* if the corresponding set covering polyhedron has no fractional vertices, and it is *kwise intersecting* if the members don’t have a common element but every k members do. We conjecture that there is a constant k such that every kwise intersecting clutter is nonideal. I will show how this conjecture for k=4 would be an extension of Jaeger’s 8flow theorem, and how a variation of the conjecture for k=3 would be an extension of Tutte’s 4flow conjecture. I will also discuss connections to tangential 2blocks, binary projective geometries, the sums of circuits property, etc. 
April 25th

JeanPhilippe Richard
Industrial and Systems Engineering University of Minnesota 
On the Convexification of Permutationinvariant Sets arising in MINLP and Data Problems
Abstract: Permutationinvariant sets  sets that do not change when the variables are permuted  appear in a variety of optimization problems, including sparse principal component analysis. Solving permutationinvariant optimization problems to global optimality requires the derivation of strong convex relaxations for their feasible region. In this talk, we study how to construct the convex hull of permutationinvariant sets in a higher dimensional space. When projection is possible, we also obtain convex hull descriptions in the original space of variables. We illustrate our techniques by developing (1) a novel reformulation and relaxation for sparse principal component analysis, (2) convex hull of matrices with bounded rank and spectral norms, (3) convex envelopes of multilinear functions over certain domains, and (4) linear descriptions of logical and sparsity constraints arising in the formulation of 01 mixed integer programs. This is joint work with Mohit Tawarmalani (Purdue) and Jinhak Kim (University of South Alabama). 
April 5th
*Friday, 1:30* *RCB 6125* (Burnaby) 
Liam Erdos, Ben Gregson, Shane Jace and Martin Zhu
Simon Fraser University 
Math 402W Operations Research Clinic
project presentation
Coordinating Primary Care Operating Hours to Reduce Acute Care Visits 
February 28th
*Thursday* *9:3012:00* *SUR 3040* 
Xiaorui Li
Ph.D. thesis defence Senior Supervisor: Z. Lu 
Sparse and Low Rank Approximation via Partial Regularization: Models, Theory and Algorithms
Abstract: Sparse representation and lowrank approximation are fundamental tools in fields of signal processing and pattern analysis. In this thesis, we consider introducing some partial regularizers to these problems in order to neutralize the bias incurred by some large entries (in magnitude) of the associated vector or some large singular values of the associated matrix. In particular, we first consider a class of constrained optimization problems whose constraints involve a cardinality or rank constraint. Under some suitable assumptions, we show that the penalty formulation based on a partial regularization is an exact reformulation of the original problem in the sense that they both share the same global minimizers. We also show that a local minimizer of the original problem is that of the penalty reformulation. Specifically, we propose a class of models with partial regularization for recovering a sparse solution of a linear system. We then study some theoretical properties of these models including existence of optimal solutions, sparsity inducing, local or global recovery and stable recovery. In addition, numerical algorithms are proposed for solving those models, in which each subproblem is solved by a nonmonotone proximal gradient (NPG) method. Despite the complication of the partial regularizers, we show that each proximal subproblem in NPG can be solved as a certain number of onedimensional optimization problems, which usually have a closedform solution. The global convergence of these methods are also established. Finally, we compare the performance of our approach with some existing approaches on both randomly generated and reallife instances, and report some promising computational results. 
PIMS  SFU Seminar
December 6th *SUR 5380* 
Asia Ivic Weiss
York University 
Regular Polyhedra, Polytopes and Beyond
Abstract: In this talk we summarize the classification of regular polyhedra and polytopes and extend the concept to that of hypertope: a thin, residually connected incidence geometry. We present the characterization of the automorphism groups of regular hypertopes and overview recent results on classification of toroidal hypertopes. 
October 6th
*Saturday, 8:304:00* *Kelowna* 
WCOM
Hosted by UBC Okanagan 
Details of the Fall 2018
West Coast Optimization Meeting
available from the hosts.

PIMS  SFU Seminar
October 4th 
JongShi Pang
Daniel J. Epstein Department of Industrial and Systems Engineering University of Southern California 
Nonproblems in Optimization for Statistics
Abstract: The phrase "nonproblems" in the title refers to a collection of adjectives that start with the prefix "non". These include "nonconvex", "nondifferentiable", "nontraditional", "nontrivial", and "nonstochastic gradient" (as a counter to a topical research theme), all in the context of optimization for statistics. Outside a standalone optimization problem, the phrase could include "noncooperative" game theory as this is an area where there are significant research opportunities when combined with the other "nonsubjects". I will present a variety of these nonproblems and give a brief summary of my research on them. 