Postdoctoral Associate Positoin at Stony Brook University

Campus Description: Stony Brook University, home to many highly ranked graduate research programs, is located 60 miles from New York City on Long Island's scenic North Shore. Our 1,100-acre campus is home to 24,000 undergraduate, graduate, and doctoral students and more than 13,500 faculty and staff, including those employed at Stony Brook University Medical Center, Suffolk County's only academic medical center and tertiary care provider. The University is a member of the prestigious Association of American Universities and co-manager of nearby Brookhaven National Laboratory (BNL), a multidisciplinary research laboratory supporting world class scientific programs utilizing state-of-the-art facilities such as the Relativistic Heavy Ion Collider, the National Synchrotron Light Source, and the Center for Functional Nanomaterials, and the NewYorkBlue IBM BG/L+P supercomputer, owned by Stony Brook and managed by BNL. Stony Brook is a partner in managing the Laboratory for the Department of Energy, and is the largest institutional scientific user of BNL facilities. As such, many opportunities exist for collaborative research, and in some cases, joint appointments can be arranged.

FAST TRACK

Descriptive Title: Postdoctoral Associate	REF#: WC-S-6779-11-06-F
Budget Title: same as above	State Professional Position
Department: Applied Math & Statistics	State Line#: 24211
Grade: SL-1	Campus: Stony Brook West Campus/HSC
	Salary: $36, 000-$46, 600

Required Qualifications: Ph.D in operations research or related discipline. Research areas to include mathematical programming, nonlinear optimization, stochastic optimization, combinatorical optimization, and applications of optimization.
Preferred Qualifications: A record of achievement to include published papers, an established reputation for research excellence, visibility within the research community, and experience writing proposals and working on funded research projects. The ideal candidate will have a strong potential for developing an externally funded research program.

Brief Description of Duties: Under the direction of the principal coordinator, the postdoctoral associate will setup, execute, and analyze computer-based simulations targeting biological systems involved in human disease such as HIV/AIDS, cancer, influenza, and botulism.

• Conduct research to include all areas of mathematical programming, optimization, modeling and simulation.
• Participate in collaborative research with faculty involved with Stony Brook Advanced Energy Research & Technology Center (AERTC) in areas of renewable energy research, power production and distribution, and system optimization for Smart Grid technology. May include areas of transportation optimization, sensor networks, wireless information technology, and quantitative finance.
• Teaching undergraduate and/or graduate courses.

Special Notes: This is a full time temporary one year appointment. FLSA Exempt position, not eligible for the overtime provisions of the FLSA. Internal and external search to occur simultaneously.

Application Procedure: Those interested in this position should submit a State employment application, cover letter and resume to:

Janice Hackney
Postdoctoral Associate (6779) Search
Applied Math and Statistics
Math Tower, P140
Stony Brook University
Stony Brook , NY 11794-3600
Fax # 631- 632-8490

Postdoctoral Position for Numerical Optimization at University of Rochester

Department of Biostatistics and Computational Biology, University of Rochester School of Medicine and Dentistry has an immediate opening for a postdoctoral position in numerical optimization. We are looking for a creative and collaborative individual to join our dynamic research group to solve the complex optimization problems in estimating parameters in large nonlinear dynamic systems. The applicant should have a Ph.D. in Applied Mathematics, Operations Research, Computer Science, Computational Engineering, or other related scientific disciplines, with theoretical and practical experience in computational programming and optimization algorithm development. Expertise and background in nonlinear and high-dimensional optimization problems for complex systems with biomedical applications is desirable. Programming skills and experience of parallel computing using OpenMP, MPI and GPU will be highly appreciated. Demonstrated evidence of good computational skills and capability of methodological/theoretical research is essential. The postdoctoral appointment is one year with possible renewal for more years subject to performance and funding availability. The scientific environment is conducive to professional development for those anticipating a career in both academic setting and applied industrial setting. This is a unique opportunity for well-trained individuals interested in both methodological and applied research. University of Rochester is an AA/EOE employer. Applications from women and minorities are encouraged to apply. Please send your CV, three references and other supporting materials to

Dr. Hulin Wu
C/O: Ms. Susan DiVincenzo (Job Code: Math-postdoc)
Department of Biostatistics and Computational Biology,
University of Rochester School of Medicine and Dentistry,
601 Elmwood Avenue, Box 630,
Rochester, NY14642.
Email: Susan_DiVincenzo@urmc.rochester.edu

Research Positions in high-performance computing in energy systems at Princeton University

Warren Powell at Princeton University is looking to fill several positions in the broad area of models and algorithms for energy systems analysis in the PENSA laboratory at Princeton University.  He is especially interested in identifying someone with strong skills and interests in high-performance computing to do research in large scale models on supercomputers.

Candidates should have a Ph.D. in operations research, computer science, applied mathematics, or physics, or a Ph.D. in a related engineering field (e.g. electrical engineering) with strong research interests in optimization models and algorithms, and their implementation in high performance computing environments.  The positions are available now, and he is hoping to fill them before the end of the calendar year.  Experience with supercomputers is valued, but not necessary.  However, he is looking for people with a research track record and enthusiasm for computational research.  Their work is primarily in the area of stochastic optimization, and while skills in stochastic optimization are useful, they are not a prerequisite for the position.

The successful candidate(s) will join a growing laboratory of faculty and professional research staff, as well as graduate and undergraduate students, in the Department of Operations Research and Financial Engineering at Princeton University.  PENSA (see http://energysystems.princeton.edu) specializes in the development of models and algorithms for a wide range of stochastic optimization problems arising in energy systems.

Interested candidates should email Warren Powell <powell@princeton.edu> with a c.v. (you are welcome to attach research articles).  After reviewing the application material, he will contact selected candidates with additional information for applying on the university website.  Please forward this email to people who may be interested.

Warren Powell
Professor and director of PENSA
Department of Operations Research and Financial Engineering
Princeton University
http://energysystems.princeton.edu

NSF AWARD:Polynomial Optimization and Convex Algebraic Geometry

Recently, NSF funded a project lead by Rekha R. Thomas on polynomial optimization and convex algebraic geometry.

This study focuses on problems from polynomial optimization and convex algebraic geometry. The latter is a new research area that concerns convex sets and convex hulls of sets that are described algebraically and arise in optimization. The key tool is the use of efficient algorithms in semidefinite programming, a branch of convex optimization that is used in polynomial optimization. The first set of questions studies the general phenomenon of when a given convex body is the linear projection of a slice (by an affine plane) of a closed convex cone. This phenomena is central to all lift-and-project methods for discrete and polynomial optimization. This study provides a uniform view of all lift-and-project methods via new notions of cone factorizations of certain operators associated to the convex body. The investigator and collaborators have recently constructed a new hierarchy of convex relaxations for algebraic sets called theta bodies. Various open questions about these bodies are posed. The methods from polynomial optimization and convex algebraic geometry can be applied to problems from computer vision. Here the application is primarily to object reconstruction from images taken by multiple cameras.

The work of the PI with her collaborators improves our understanding of the algebraic and geometric structures that underlie optimization problems that involve polynomials. Such problems have a wide array of applications and admit methods from both the algebraic and analytic sides of mathematics. This research considers both improvements in our understanding of the theoretical aspects of polynomial optimization, and the application of these methods to problems in computer vision.

NSF AWARD:Interior-point algorithms for conic optimization with sparse matrix cone constraints

Recently NSF funded a project lead by Lieven Vandenberghe on interior-point algorithms for conic optimization with sparse matrix cone constraints. 

Conic optimization is an extension of linear programming in which the componentwise vector inequalities are replaced by inequalities with respect to nonpolyhedral convex cones. The conic optimization model is widely used in the recent literature on convex optimization and provides an elegant framework for extending interior-point algorithms from linear programming to convex optimization. It is also the basis of popular modeling systems for convex optimization.

The research on algorithms for conic optimization has mainly focused on three types of inequalities, associated with the nonnegative orthant, the second-order cone, and the positive semidefinite cone. This restriction is motivated by symmetry properties that can be exploited to formulate symmetric primal-dual interior-point algorithms. However, large gaps in linear algebra complexity exist between the three types of conic constraints, and this can lead to inefficiencies when convex optimization problems are converted to the standard conic format.

This study considers approaches to improve the efficiency of conic optimization solvers by considering a larger class of conic constraints, defined by chordal sparse matrix cones, i.e., cones of positive semidefinite matrices with a given chordal sparsity pattern, and the associated dual cones of chordal sparse matrices that have a positive semidefinite completion. These cones include as special cases the three standard cones, but also several interesting non-self-dual cones. Moreover non-chordal sparsity patterns can often be efficiently embedded in chordal patterns and, as a consequence, sparse semidefinite programs can be solved as non-symmetric cone programs involving lower-dimensional cones than the positive semidefinite cone used in semidefinite programming methods. The choice for chordal matrix cones is further motivated by the existence of fast algorithms for evaluating the associated barrier functions and their derivatives.

The investigator and his collaborators study nonsymmetric interior-point algorithms for sparse matrix cones, building on techniques developed for large-scale sparse matrix computations, in particular, multifrontal and supernodal factorization algorithms and parallel sparse matrix algorithms. A wide variety of practical problems in engineering and science can be formulated as nonlinear convex optimization problems, and solved using algorithms developed over the last few decades.

The success of these techniques has created a demand for robust and efficient algorithms for very large convex optimization problems, especially for applications in machine learning, computer vision, electronic design automation, sensor networks, and combinatorial optimization. The problem sizes that arise in these fields often exceed the capabilities of general-purpose solvers. The work of the prinicipal investigator with his collaborators considers approaches to improve the scalability of interior-point algorithms, an important class of convex optimization algorithms. Freely available high-quality software implementations of the techniques developed in the project are a product of the research.

NSF AWARD:Ranking and Clustering by Integer and Linear Optimization

Recently, NSF funded a project lead by Amy N.Langville on ranking and clustering by integer and linear optimization. 

The research component of this proposal is about ranking and clustering. A given collection of items is to be ordered according to some criterion (e.g., from most to least important). In clustering, the goal is to group items so that similar items appear together. Though well-studied, ranking and clustering research has been dominated by heuristic methods that produce approximate results quickly. Optimization methods that produce exact optimal results have been far less studied, possibly because these methods require longer computational times, and the gains in accuracy do not outweigh the computational costs in many applications. This is unfortunate since optimization methods are based on a beautiful theoretical framework and can lead to novel and interesting results. For example, preliminary work by the PI on a ranking optimization method, indicates that multiple optimal solutions can be linked to ties in the ranked list. On the other hand, heuristic methods are not designed to handle ties in the output ranking. Two main goals of the proposed research are: (1) to reveal interesting theoretical connections, and (2) to increase the size limits of optimally solvable problems using both classical and clever new relaxation techniques.

Ranking, also known as linear ordering (LOR), is close in spirit to the Traveling Salesman Problem (TSP): both are simple to state, yet hard to solve optimally. Over several decades, huge gains have been made on the size of solvable TSPs, that have resulted in new and unforeseen uses. Notable examples occur in the transportation industry (involving thousand-city TSPs) and in the microprocessor industry (with even larger TSPs routing copper wiring on circuit-boards). Similar progress is expected for the LOR, whose current limit is a few hundred to a few thousand items in some cases. Yet applications for much larger LORs abound, e.g. rankings of genes, products, and webpages. Furthermore, breakthroughs for the TSP (and likewise the proposed LOR work) have not solely been related to scale, but theoretical advances and progress in understanding connections to other problems and fields have been equally crucial in the development of general purpose methods for integer programming, such as cutting plane and branch and cut techniques.

The proposed research has great impact. Ranking and clustering have become standard data analysis tools with many uses. For example, Google uses ranking to order webpages resulting from user queries, and also uses clustering for its "Find Similar Pages" function. Amazon uses clustering in its "Customers Who Bought X Also Bought Y" feature. Facebook and Twitter can generate ads and friend recommendations based on ranking and clustering algorithms. To ensure the results of the proposed research reach these consumers, the work will be widely disseminated through journal publications and two books. The project's novel student training plan, which includes peer mentoring and an international exchange program, will broaden participation of underrepresented groups. The educational component and its Calculus Activity Book, being aimed at changing students attitudes towards the sciences and at instilling confidence in scientific ability, will have the stronger impact on those students who more commonly fail to be retained (likely a relatively large number coming from underrepresented groups), and thus will help further diversify and broaden participation in STEM disciplines.

Seminar: Valid Polynomial Inequality Generation in Polynomial Optimization

SPEAKER:               Professor Miguel F. Anjos
                                    Mathematics and Industrial Engineering
                                    Ecole Polytechnique de Montreal, Canada
 
DATE / TIME:          Monday, June 13, 2011
                                    2:00 – 3:00 p.m.
 
LOCATION:             Room 453 Mohler Lab, 200 W. Packer Avenue
 
 
ABSTRACT:  Polynomial optimization problems are normally solved using hierarchies of convex relaxations.  These schemes rapidly become computationally expensive, and are usually tractable only for problems of small sizes.  We propose a novel dynamic scheme for generating valid polynomial inequalities for general polynomial optimization problems.  These valid inequalities are then used to construct better approximations of the original problem.  For the special case of binary polynomial problems, we prove that the proposed scheme converges to the global optimal solution under mild assumptions on the initial approximation of the problem.  We also present examples illustrating the computational behavior of the scheme and compare it to other methods in the literature.
 
This is joint work with Bissan Ghaddar (U of Waterloo) and Juan Vera (Tilburg U).
 

Post-doc position at Division of Systems and Engineering Management, Nanyang Technological University

A postdoctoral research scholar position is available at Singapore Nanyang Technological University (NTU) to reduce the excess production of the active pharmaceutical Ingredient. This position is supported by a three year grant from GlaxoSmithKline, a leading international healthcare company. Potential candidates should be familiar with inventory models and supply chain coordination as well as strong motivation in solving practical problems in pharmaceutical industry. The initial appointment will be for one year, with additional years contingent upon mutual agreement.

Applicants must be capable of conducting projects independently, writing manuscripts, and presenting their research at international meetings. The ideal applicant should be highly motivated, able to interact with other people and have expertise in supply chain management. A PhD in a relevant field and a publication record is required. Fluency in English is essential.

Requirements:

1) Interested in solving challenging problems in pharmaceutical industry

2) PhD degree in operations research or relevant fields

3) Knowledge of supply chain management

4) Knowledge of queueing theory (preferred)

5) Simulation (preferred)

6) Matlab (preferred)

This is a full time job at NTU in Singapore. The monthly salary is competitive with up to two month annual bonus. This is a great opportunity for someone who wants to expose himself to the pharmaceutical industry and experience the life in Singapore. If interested, please send your resume, transcripts and a brief biography to Dr. Kan Wu (wukan@ntu.edu.sg).