INDUSTRIAL AND SYSTEMS ENGINEERING DEPARTMENT
SPRING 2011 SEMINAR SERIES
TITLE: Mixed-Integer Programming Models and Methods for Optimization under Joint Chance Constraints
SPEAKER: Dr. Simge Küçükyavuz, Assistant Professor in the Integrated Systems Engineering Department, Ohio State University.
DATE / TIME: Wednesday, March 2, 2011
2:30 – 3:30 p.m.
LOCATION: Room 453 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: In this talk, we consider a class of stochastic optimization problems with quality of service or reliability constraints. These problems occur frequently in practice; and they are dynamic, contain uncertain data, and involve discrete decisions. The resulting multi-stage stochastic mixed-integer programs are challenging both theoretically and computationally. The service level restrictions are modeled with joint chance constraints, which are non-convex. In addition, the deterministic equivalents of these chance-constrained optimization problems are very large-scale mixed-integer programs. We first review two formulations for chance constrained linear programs. One is based on the so-called mixing set reformulation, and another is based on disjunctive programming. We give valid inequalities that strengthen the mixing set reformulation. We also prove that the disjunctive programming formulation is compact and tight under certain conditions. We summarize our computational experiments with both the static and the dynamic probabilistic lot-sizing problems to illustrate the effectiveness of a branch-and-cut method using the proposed inequalities.
BIOGRAPHY: Simge Küçükyavuz is an Assistant Professor in the Integrated Systems Engineering Department at the Ohio State University. Her research interests are in mixed-integer and combinatorial optimization, optimization under uncertainty and their applications. She received a PhD and MS in Operations Research from the University of California at Berkeley, and a BS in Industrial Engineering from Middle East Technical University, Turkey. Prior to joining Ohio State in 2009, she was an Assistant Professor at the University of Arizona and a Research Associate at Hewlett-Packard Laboratories. Her research has been supported by the National Science Foundation
.
Monday, February 28, 2011
[Optimization-society] Post Doctoral Position in Optimization, Mathematical Sciences, Chalmers University of Technology, Gothenburg, Sweden
Post Doctoral Position in Optimization with Application to The Maintenance Planning Optimization for Wind Turbines and Farms at the Department of Mathematical Sciences, Chalmers
Reference number 2011/52
Application deadline 2011-04-15
Job start date 2011-07-01
ENERGY – a Chalmers Area of Advance
We are standing on the threshold of radical changes in European and global energy systems. Chalmers' energy-related research is at the heart of both the challenges and the opportunities presented to industry and society by these changes. Conducting world class energy technology and systems research is fundamental within the Energy Area of Advance. Special strengths are the close cooperations between different energy technology and system areas and the development of internationally acclaimed energy system models.
Chalmers Energy Initiative is a large strategic research programme within the Energy Area of Advance, covering four research areas: Energy Combines, Electricity propulsion systems and hybrid vehicles, Large-scale renewable electricity generation and grid integration, and Technology Impact Assessment.
The Department of Mathematical Sciences houses about 80 researchers active in a wide range of research fields in an internationally recognized, strong research environment. The department belongs to both Chalmers University of Technology and the University of Gothenburg. You can find information about our research groups at http://www.chalmers.se/math/EN/research
The postdoctoral position will be placed within the optimization group. It currently hosts one full professor, one associate professor, one adjunct professor, one postdoc, and four PhD students. The research performed within the group spans theoretical as well as applied and industrial projects, with a current emphasis on problems of a combinatorial and integer optimization nature.
We are standing on the threshold of radical changes in European and global energy systems. Chalmers' energy-related research is at the heart of both the challenges and the opportunities presented to industry and society by these changes. Conducting world class energy technology and systems research is fundamental within the Energy Area of Advance. Special strengths are the close cooperations between different energy technology and system areas and the development of internationally acclaimed energy system models.
Chalmers Energy Initiative is a large strategic research programme within the Energy Area of Advance, covering four research areas: Energy Combines, Electricity propulsion systems and hybrid vehicles, Large-scale renewable electricity generation and grid integration, and Technology Impact Assessment.
The Department of Mathematical Sciences houses about 80 researchers active in a wide range of research fields in an internationally recognized, strong research environment. The department belongs to both Chalmers University of Technology and the University of Gothenburg. You can find information about our research groups at http://www.chalmers.se/math/EN/research
The postdoctoral position will be placed within the optimization group. It currently hosts one full professor, one associate professor, one adjunct professor, one postdoc, and four PhD students. The research performed within the group spans theoretical as well as applied and industrial projects, with a current emphasis on problems of a combinatorial and integer optimization nature.
Job description
The Post Doc position is placed within the field of maintenance planning optimization. Current research topics include mathematical optimization models from the points of view of complexity, polyhedral theory, and the efficient computability of (near-)optimal maintenance schedules, including the modelling of the uncertainty of parameter values such as the remaining lives of components. The current position is aimed at placing special emphasis on the construction and study of maintenance optimization models in the energy intensive and power production industries, in particular the wind power sector. Part of the research will be performed in close collaboration with researchers at the Division of Electric Power Engineering at Chalmers.
Link to Electric Power Engineering: http://www.chalmers.se/ee/EN/research/research-divisions/epe
A Post Doc position is an appointment that offers an opportunity to qualify for further research positions within academia or industry. The appointment is for 1+1 years of full-time employment. The majority of your working time is devoted to your research. Included in your work is also the co-supervision of Ph.D. students and M.Sc. thesis students, as well as a limited contribution to the teaching of graduate students. In particular, you will be expected to take part in the supervision of a Ph.D. student devoted to work on maintenance optimization who is currently being recruited. You will attend international conferences and meetings, where you will present your own work and establish useful contacts with other researchers. The position will commence July 1, 2011.
More details about the research performed in the optimization group can be found at http://www.chalmers.se/math/EN/research/research-groups/optimization
The Postdoc position is an appointment that offers an opportunity to qualify for further research positions within academia or industry. The appointment is for 1+1 years of full-time employment, with a monthly salary paid. Most of your working time is devoted to your own research. Included in your work is also to take part in supervision of Ph.D. students and M.Sc thesis students. Teaching of undergraduate students may also be included to a small extent. You will attend international conferences and meetings, where you will present your own work and establish useful contacts with other researchers. The position can commence immediately.
Link to Electric Power Engineering: http://www.chalmers.se/ee/EN/research/research-divisions/epe
A Post Doc position is an appointment that offers an opportunity to qualify for further research positions within academia or industry. The appointment is for 1+1 years of full-time employment. The majority of your working time is devoted to your research. Included in your work is also the co-supervision of Ph.D. students and M.Sc. thesis students, as well as a limited contribution to the teaching of graduate students. In particular, you will be expected to take part in the supervision of a Ph.D. student devoted to work on maintenance optimization who is currently being recruited. You will attend international conferences and meetings, where you will present your own work and establish useful contacts with other researchers. The position will commence July 1, 2011.
More details about the research performed in the optimization group can be found at http://www.chalmers.se/math/EN/research/research-groups/optimization
The Postdoc position is an appointment that offers an opportunity to qualify for further research positions within academia or industry. The appointment is for 1+1 years of full-time employment, with a monthly salary paid. Most of your working time is devoted to your own research. Included in your work is also to take part in supervision of Ph.D. students and M.Sc thesis students. Teaching of undergraduate students may also be included to a small extent. You will attend international conferences and meetings, where you will present your own work and establish useful contacts with other researchers. The position can commence immediately.
Required qualifications
The applicant should have a recent Ph.D. degree in mathematics, applied mathematics, or computer science, with strength in mathematical optimization. Experience in combinatorial optimization is highly desirable. You have a good theoretical background and a strong interest in working on real-world problems in interdisciplinary projects, with researchers and engineers from other research fields.
You must have a strong track record in research within the field, demonstrated by scientific publications, etc. Excellent interpersonal, and communication skills are required. Hence, excellent English skills are required, in oral as well as writing communication. An good command of Swedish is strongly encouraged.
You must have a strong track record in research within the field, demonstrated by scientific publications, etc. Excellent interpersonal, and communication skills are required. Hence, excellent English skills are required, in oral as well as writing communication. An good command of Swedish is strongly encouraged.
Application procedure
The application shall be written in English and include the following items:
The documents according to items 1-4 above shall be attached as two pdf-files.
One should contain items 1-3 in the listing of material to be included in the application The other should contain item 4 in the listing of material, and any other appendices.The files may be compressed (zipped).
If any material is not available electronically or cannot be transferred to pdf format, the material can be sent as a hard copy to Registrar. The applicants name and the reference number (2011/52) must be written on the first page of the application.
Address:
Registrar
Chalmers University of Technology
SE-412 96 Göteborg
Sweden
- An application of a maximum of one A4 page containing your specific qualifications for the position
- Curriculum Vitae including list of publications
- Two reference persons who can be contacted by Chalmers (describe association with them and give their contact addresses)
- Attested copies of education certificates, including grade reports and other documents
The documents according to items 1-4 above shall be attached as two pdf-files.
One should contain items 1-3 in the listing of material to be included in the application The other should contain item 4 in the listing of material, and any other appendices.The files may be compressed (zipped).
If any material is not available electronically or cannot be transferred to pdf format, the material can be sent as a hard copy to Registrar. The applicants name and the reference number (2011/52) must be written on the first page of the application.
Address:
Registrar
Chalmers University of Technology
SE-412 96 Göteborg
Sweden
Further information
In addition to what is stated above please include the following to your application:
- Two recommendation letters (marked with the reference number 2011/52)
- Copies of your two most relevant scientific publications
- Transcripts (grades) and diplomas from Master’s and Ph.D. programs
Further information:
Professor Michael Patriksson
Mathematical Sciences, Chalmers
E-mail: mipat@chalmers.se
Phone: +46 31 772 3529
- Two recommendation letters (marked with the reference number 2011/52)
- Copies of your two most relevant scientific publications
- Transcripts (grades) and diplomas from Master’s and Ph.D. programs
Further information:
Professor Michael Patriksson
Mathematical Sciences, Chalmers
E-mail: mipat@chalmers.se
Phone: +46 31 772 3529
Friday, February 25, 2011
ISE seminar 2-22-11
SPRING 2011 SEMINAR SERIES
TITLE: Polytopes and Arrangements: Diameter and Curvature
SPEAKER: Dr. Yuriy Zinchenko, Assistant Professor of Mathematics
University of Calgary, Canada
DATE / TIME: Tuesday, February 22, 2011
1:30 – 2:30 p.m.
LOCATION: Room 375 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: The curvature of a polytope, defined as the largest possible total curvature of the associated central path, can be regarded as a continuous analogue of its diameter. Algorithmically, the diameter may be interpreted as an informal guideline for estimating the complexity of solving a particular linear optimization problem with simplex method. Similarly, the curvature may be thought of as a predictor for the efficiency of path-following interior-point methods on a given problem instance. We introduce a continuous analogue of the Hirsch conjecture and a discrete analogue of the result of Dedieu, Malajovich and Shub. We prove a continuous analogue of the result of Holt and Klee, namely, we construct a family of polytopes, which attain the conjectured order of the largest curvature. We prove the analogue of the result of Klee and Walkup. Namely, we show that if the order of the curvature is less than the dimension d for all polytopes defined by 2d inequalities for all d, then the order of the curvature is less then the number of defining inequalities for all polytopes.
BIOGRAPHY: Dr. Yuriy Zinchenko is an Assistant Professor of Mathematics at the University of Calgary. He received his Ph.D. in Operations Research from Cornell University in 2005. In 2005-2008 Dr. Zinchenko held a post-doctoral position at the Advanced Optimization Laboratory, McMaster University, and in 2006 - 2008 was also a post-doctoral researcher with the Department of Radiation Oncology, Princess Margaret Hospital. In 2007 Dr. Zinchenko was a recipient of 2007 MITACS Award for Best Novel Use of Mathematics in Technology Transfer for his work in optimal and robust radiotherapy design. Dr. Zinchenko’s research interests include operations research, optimization algorithms and software, large-scale computational optimization, applications to medicine and health care, particularly, optimal radiation therapy design, scientific parallel computing and high-performance linear algebra.
TITLE: Polytopes and Arrangements: Diameter and Curvature
SPEAKER: Dr. Yuriy Zinchenko, Assistant Professor of Mathematics
University of Calgary, Canada
DATE / TIME: Tuesday, February 22, 2011
1:30 – 2:30 p.m.
LOCATION: Room 375 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: The curvature of a polytope, defined as the largest possible total curvature of the associated central path, can be regarded as a continuous analogue of its diameter. Algorithmically, the diameter may be interpreted as an informal guideline for estimating the complexity of solving a particular linear optimization problem with simplex method. Similarly, the curvature may be thought of as a predictor for the efficiency of path-following interior-point methods on a given problem instance. We introduce a continuous analogue of the Hirsch conjecture and a discrete analogue of the result of Dedieu, Malajovich and Shub. We prove a continuous analogue of the result of Holt and Klee, namely, we construct a family of polytopes, which attain the conjectured order of the largest curvature. We prove the analogue of the result of Klee and Walkup. Namely, we show that if the order of the curvature is less than the dimension d for all polytopes defined by 2d inequalities for all d, then the order of the curvature is less then the number of defining inequalities for all polytopes.
BIOGRAPHY: Dr. Yuriy Zinchenko is an Assistant Professor of Mathematics at the University of Calgary. He received his Ph.D. in Operations Research from Cornell University in 2005. In 2005-2008 Dr. Zinchenko held a post-doctoral position at the Advanced Optimization Laboratory, McMaster University, and in 2006 - 2008 was also a post-doctoral researcher with the Department of Radiation Oncology, Princess Margaret Hospital. In 2007 Dr. Zinchenko was a recipient of 2007 MITACS Award for Best Novel Use of Mathematics in Technology Transfer for his work in optimal and robust radiotherapy design. Dr. Zinchenko’s research interests include operations research, optimization algorithms and software, large-scale computational optimization, applications to medicine and health care, particularly, optimal radiation therapy design, scientific parallel computing and high-performance linear algebra.
OR Analyst and Software Developer @Oliver Wyman
Position Overview
Oliver Wyman is seeking candidates with Operations Research and software development skills to join its transportation software development team, based in Princeton, New Jersey. The position will design and develop new software, as well as maintain existing software, that uses many algorithms for optimizing transport and transportation related operations. Our practice is a global leader in railroad operations planning applications and consulting, and among much recognition our group received the Franz Edelman prize for the practice of management science in 2003. Our development team is responsible for designing, building, and maintaining the MultiRail family of products, which is the global standard platform for railway service and operations planning. We are seeking an individual who is capable of working as an integral member of our approximately 12 person development team.
About Oliver Wyman
As one of the world’s premier corporate strategy and operations firms, Oliver Wyman helps leading enterprises develop, build, and operate strong businesses that deliver sustained shareholder value growth. Oliver Wyman’s proprietary business design techniques, combined with its specialized industry knowledge and global reach, enable companies to anticipate changes in customer priorities and the competitive environment, and then design their businesses and improve operations to seize opportunities created by those changes. The firm serves clients from 22 offices in the Americas, Europe, and Asia.
OR Analyst and Software Developer
Job Description
Description
* Design and develop new software, including the development of algorithms, as well as maintain existing software and algorithms, using C#, C++ and SQL
Qualifications
* Experience with object oriented development in C#, C++ or some other OO language
* Strong background and interest in optimization (linear, integer, combinatorial) and heuristics
* The ability to work well with our team
* Eagerness to work on challenging problems
* Degree in Operations Research, Computer Science or equivalent, or 2+ years of relevant work experience
Plusses:
* Experience with optimization algorithms for transportation problems
* Understanding of advanced statistical methods
* Experience with mathematical modeling languages such as AMPL
* Experience with statistical languages such as R
* Experience with a simulation language such as GPSS
* Some experience in databases, SQL, MS Access, VBA
* Windows and Linux experience
Finding multiple optimal solutions of a BILP
It's possible to get CPLEX to return more than one optimal objective value,
using the "solution pool" feature described on page 71 of the CPLEX-for-AMPL
user guide, www.ampl.com/BOOKLETS/
CPLEX gives you the incumbent feasible solutions that it found, however,
only one of which is optimal. I have found the following settings useful
for generating multiple solutions all having the optimal objective value:
option cplex_options 'poolstub=solfile populate=1 poolintensity=4
poolagap=0';
You might want to experiment with other settings as well. As explained in
the user guide, you will need to write an AMPL "for" loop to retrieve the
different solutions one by one for display or processing.
using the "solution pool" feature described on page 71 of the CPLEX-for-AMPL
user guide, www.ampl.com/BOOKLETS/
CPLEX gives you the incumbent feasible solutions that it found, however,
only one of which is optimal. I have found the following settings useful
for generating multiple solutions all having the optimal objective value:
option cplex_options 'poolstub=solfile populate=1 poolintensity=4
poolagap=0';
You might want to experiment with other settings as well. As explained in
the user guide, you will need to write an AMPL "for" loop to retrieve the
different solutions one by one for display or processing.
Thursday, February 24, 2011
[Msom-society] Post-Doctoral Researcher Position at EPFL
Post-Doctoral Researcher Position
the Swiss Federal Institute of Technology in Lausanne (EPFL)
We are offering a challenging position as a full-time post-doctoral researcher within the College of Management of Technology at the Swiss Federal Institute of Technology in Lausanne (EPFL).
The TOM chair is focused on Technology & Operations Management. Motivated by real-world applications, we base our work on stylized mathematical models in order to capture essential tradeoffs and offer managerial insights in decision-making. We also conduct survey based work and strongly encourage industry collaboration to help bridge between theory and practice.
In recent years, we have been interested in various topics related to supply chain management: its connection with finance, the impact of environmental decisions, or the alignment between supply chains and product portfolios. Other subjects include the assortment strategy in retail stores, the coordination in new product introductions, or the management of technology standards. Please visit http://tom.epfl.ch for further information.
In accordance with the chair’s research interests, two alternative profiles are targeted for the post-doctoral position. The first one has a strong mathematical background with a proven interest for applications in operations management. The second one has strong background in statistics and survey analysis with an interest in supply chain and supply chain finance related applications.
We expect
· High level of motivation for academic research work motivated by relevant practical problems.
· Excellent research skills demonstrated by a good publication record.
· Proficiency in English (working language), strong team spirit, social skills and independence.
First profile
· An excellent PhD degree, preferably in Operations Management/Research, Industrial Engineering, or Mathematics.
· A strong mathematical background and a sincere interest in business-related topics.
Second profile
· An excellent PhD degree, preferably in Operations Management/Research, Statistics or Finance.
· Interest in Finance, and its interface with Supply Chain Management, as well as interest in Statistics (database and/or survey analysis).
We offer
· Excellent working conditions and collaboration with leading academics.
· Active involvement in ongoing topics within our research team.
· Opportunity to advance own research.
· Full financial support.
Contact
Please send your application file (cover letter, resume, representative publications) to Jean-Sébastien Tancrez (jean-sebastien.tancrez@epfl.
Teacher Development Series SPRING 2011
Session 1 - How Students Learn: Effective Assessment
When: Thursday February 10th, 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: How does one effectively assess student learning? What are the roles of quizzes, tests, essays and other forms of assessment in the learning process? How can effective assessment help improve teaching?
Who: Greg Reihman (Faculty Development/CAS-Philosophy).
Session 2 – How Teachers Teach: Writing to Learn
When: Thursday February 24th, 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: How can teachers use shorter, informal, easy-to-grade writing assignments to facilitate learning?
Who: Gregory Skutches (Writing Across the Curriculum/CAS-English)
Session 3 - How Teachers Teach: Using Technology to Enhance Learning
When: Thursday March 3rd, 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: Why use instructional technologies to enhance learning? How should such technologies be integrated into classes so they are used most effectively? What instructional technologies are available and how are they being used?
Who: Ilena Key and Judd Hark (LTS-Instructional Technology Team)
Session 4 - How Teachers Teach: Developing a Syllabus
When: Thursday 3/24 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: What is the purpose of a syllabus? What should be included? How can an effective syllabus help both instructors and students?
Who: Greg Reihman (Faculty Development/CAS-Philosophy)
Session 5 - How Students Learn: The Role of Cognition in Instructional Design
When: Thursday 4/7 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: What are our best theories about how people receive and process information? How should teachers take these theories into account when teaching undergraduates?
Who: M.J. Bishop (COE-Teaching, Learning, Technology)
Session 6- How Teachers Teach: Explanations and Investigations
When: Thursday 4/21 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: How do effective faculty explain difficult concepts? How do they help students learn which questions to ask and how to find good answers?
Who: Kristen Jellison (RCEAS - Civil & Environmental Engineering)
When: Thursday February 10th, 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: How does one effectively assess student learning? What are the roles of quizzes, tests, essays and other forms of assessment in the learning process? How can effective assessment help improve teaching?
Who: Greg Reihman (Faculty Development/CAS-Philosophy).
Session 2 – How Teachers Teach: Writing to Learn
When: Thursday February 24th, 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: How can teachers use shorter, informal, easy-to-grade writing assignments to facilitate learning?
Who: Gregory Skutches (Writing Across the Curriculum/CAS-English)
Session 3 - How Teachers Teach: Using Technology to Enhance Learning
When: Thursday March 3rd, 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: Why use instructional technologies to enhance learning? How should such technologies be integrated into classes so they are used most effectively? What instructional technologies are available and how are they being used?
Who: Ilena Key and Judd Hark (LTS-Instructional Technology Team)
Session 4 - How Teachers Teach: Developing a Syllabus
When: Thursday 3/24 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: What is the purpose of a syllabus? What should be included? How can an effective syllabus help both instructors and students?
Who: Greg Reihman (Faculty Development/CAS-Philosophy)
Session 5 - How Students Learn: The Role of Cognition in Instructional Design
When: Thursday 4/7 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: What are our best theories about how people receive and process information? How should teachers take these theories into account when teaching undergraduates?
Who: M.J. Bishop (COE-Teaching, Learning, Technology)
Session 6- How Teachers Teach: Explanations and Investigations
When: Thursday 4/21 2:30-4:00
Where: EWFM 379 (Media Center Classroom)
Topics: How do effective faculty explain difficult concepts? How do they help students learn which questions to ask and how to find good answers?
Who: Kristen Jellison (RCEAS - Civil & Environmental Engineering)
[aco-announce] ACO Seminar - Aranyak Mehta - Google Research
Aranyak Mehta, Google Research, Mountain View.
Location and Time: KACB 1116E, Thur, Feb 24, 4:30pm.
Host: V. Vazirani (please contact V.V. to meet with the speaker)
------------------------------
Online Matching and the Adwords Market
------------------------------
The spectacular success of search and display advertising -- to
businesses and search engine companies -- and its huge growth
potential has attracted the attention of researchers from many aspects
of computer science. Since a core problem in this area is that of
effective ad allocation, an inherently algorithmic and game-theoretic
question, numerous theoreticians have worked in this area in recent
years. Ad allocation involves matching ad slots to advertisers, under
demand and supply constraints. In short, the better the matching, the
more efficient the market.
Interestingly, the seminal work on online matching, by Karp, Vazirani
and Vazirani, was done over two decades ago, well before the advent of
the Internet economy! In this talk, I will give an overview of several
key algorithmic papers in this area, starting with its purely academic
beginnings, to papers that actually address the Adwords problem. The
elegant -- and deep -- theory behind these algorithms involves new
combinatorial, probabilistic and linear programming techniques.
Besides the classic KVV paper (STOC 1990), this talk will refer to
several papers with my co-authors:
Mehta, Saberi, Vazirani, Vazirani (FOCS 05, J. ACM 07),
Goel, Mehta (SODA 08),
Feldman, Mehta, Mirrokni, Muthukrishnan (FOCS 09),
Aggarwal, Goel, Karande, Mehta (SODA 10),
Karande, Mehta, Tripathi (STOC 11).
[Transci-logistics-society] Research position at SINTEF ICT
SINTEF ICT offers a 3 year temporary research position in discrete optimization methods in maritime and road-based transportation.
SINTEF ICT is an institute within the SINTEF Group, the largest independent research organization in Scandinavia.
SINTEF ICT undertakes research and development projects for industry, the public sector, and funding agencies
in the area of information and communication technology. SINTEF ICT has approximately 280 employees and
collaborates with a large number of industries and academic institutions in Norway and abroad.
The Department of Applied Mathematics at SINTEF ICT performs research activities within modeling and
simulation of petroleum reservoirs, geometric modeling, visualization, parallel and heterogeneous scientific computing,
and optimization methods for complex planning and scheduling applications.
The Department of Applied Mathematics has recently been awarded a research project ("DOMinant II")
in discrete optimization methods in maritime and road-based transportation. We are currently seeking an
innovative postdoctoral researcher with excellent skills in mathematics, operations research, and transportation optimization.
The research fellow will be working in DOMinant II. There are three main partners within the project:
- Department of Industrial Economics and Technology Management at NTNU
- Molde University College
- SINTEF ICT.
Workplace: SINTEF ICT's premises in Oslo, Norway.
The application deadline is April 1 2011. Electronic application only.
For the full call text and instructions on the electronic application procedure we refer to
http://www.sintef.no/Home/
SINTEF ICT is an institute within the SINTEF Group, the largest independent research organization in Scandinavia.
SINTEF ICT undertakes research and development projects for industry, the public sector, and funding agencies
in the area of information and communication technology. SINTEF ICT has approximately 280 employees and
collaborates with a large number of industries and academic institutions in Norway and abroad.
The Department of Applied Mathematics at SINTEF ICT performs research activities within modeling and
simulation of petroleum reservoirs, geometric modeling, visualization, parallel and heterogeneous scientific computing,
and optimization methods for complex planning and scheduling applications.
The Department of Applied Mathematics has recently been awarded a research project ("DOMinant II")
in discrete optimization methods in maritime and road-based transportation. We are currently seeking an
innovative postdoctoral researcher with excellent skills in mathematics, operations research, and transportation optimization.
The research fellow will be working in DOMinant II. There are three main partners within the project:
- Department of Industrial Economics and Technology Management at NTNU
- Molde University College
- SINTEF ICT.
Workplace: SINTEF ICT's premises in Oslo, Norway.
The application deadline is April 1 2011. Electronic application only.
For the full call text and instructions on the electronic application procedure we refer to
http://www.sintef.no/Home/
ISE seminar- Tuesday 3-1
INDUSTRIAL AND SYSTEMS ENGINEERING DEPARTMENT
SPRING 2011 SEMINAR SERIES
TITLE: Using Active-Set Phases to Accelerate Optimization Algorithms
SPEAKER: Dr. Daniel Robinson
Northwestern University, Illinois
DATE / TIME: Tuesday, March 1, 2011
1:30 – 2:30 p.m.
LOCATION: Room 375 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: In the first part of the talk I will clarify the meaning of “active-set phases” by giving an overview of three optimization algorithms that are currently being developed. The first algorithm is designed for solving linear complementarity problems and is motivated by a financial optimization problem. The second algorithm is used for solving (convex) stochastic optimization problems and will be described in the context of a speech recognition application. The third algorithm is a general-purpose second- derivative sequential quadratic programming method that is designed for solving sparse large-scale nonlinear and nonconvex problems. Since this method benefits from so-called “warm-starting”, I will motivate its development by considering a simple trajectory optimization problem.
I will dedicate the second part of the talk to a more detailed description of the sequential quadratic programming algorithm. In particular, I will (i) describe various difficulties associated with developing second-derivative sequential quadratic programming methods; (ii) explain the strategies that I used to avoid these hurdles; (iii) provide global and local convergence results; and (iv) furnish numerical tests that elucidate the benefits of an active-set phase.
BIOGRAPHY: Daniel Robinson received his Ph.D. from the Mathematics and Statistics Department at the University of California, San Diego in 2007. In the summer of 2006 he worked for Northrop Grumman as a consultant with his advisor Philip Gill and developed algorithms for trajectory optimization problems. From 2007-2010, he worked as a research assistant to Nick Gould at the University of Oxford where he developed and implemented optimization algorithms for large-scale nonlinear and convex optimization. During this time, he also maintained a visiting position at the Rutherford Appleton Laboratory, formally known as Harwell. Daniel is currently a post-doc for Jorge Nocedal in the Industrial Engineering and Management Sciences Department at Northwestern University, and his current research interests include algorithms for machine learning, linear complementarity problems, and large-scale nonlinear optimization.
ALL FULL-TIME ISE PHD STUDENTS ARE REQUIRED TO ATTEND AND
ALL ISE GRADUATE STUDENTS ARE ENCOURAGED TO ATTEND
REFRESHMENTS WILL BE SERVED FOLLOWING THE SEMINAR – IN THE 4TH FLOOR GOTT LOUNGE AREA
SPRING 2011 SEMINAR SERIES
TITLE: Using Active-Set Phases to Accelerate Optimization Algorithms
SPEAKER: Dr. Daniel Robinson
Northwestern University, Illinois
DATE / TIME: Tuesday, March 1, 2011
1:30 – 2:30 p.m.
LOCATION: Room 375 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: In the first part of the talk I will clarify the meaning of “active-set phases” by giving an overview of three optimization algorithms that are currently being developed. The first algorithm is designed for solving linear complementarity problems and is motivated by a financial optimization problem. The second algorithm is used for solving (convex) stochastic optimization problems and will be described in the context of a speech recognition application. The third algorithm is a general-purpose second- derivative sequential quadratic programming method that is designed for solving sparse large-scale nonlinear and nonconvex problems. Since this method benefits from so-called “warm-starting”, I will motivate its development by considering a simple trajectory optimization problem.
I will dedicate the second part of the talk to a more detailed description of the sequential quadratic programming algorithm. In particular, I will (i) describe various difficulties associated with developing second-derivative sequential quadratic programming methods; (ii) explain the strategies that I used to avoid these hurdles; (iii) provide global and local convergence results; and (iv) furnish numerical tests that elucidate the benefits of an active-set phase.
BIOGRAPHY: Daniel Robinson received his Ph.D. from the Mathematics and Statistics Department at the University of California, San Diego in 2007. In the summer of 2006 he worked for Northrop Grumman as a consultant with his advisor Philip Gill and developed algorithms for trajectory optimization problems. From 2007-2010, he worked as a research assistant to Nick Gould at the University of Oxford where he developed and implemented optimization algorithms for large-scale nonlinear and convex optimization. During this time, he also maintained a visiting position at the Rutherford Appleton Laboratory, formally known as Harwell. Daniel is currently a post-doc for Jorge Nocedal in the Industrial Engineering and Management Sciences Department at Northwestern University, and his current research interests include algorithms for machine learning, linear complementarity problems, and large-scale nonlinear optimization.
ALL FULL-TIME ISE PHD STUDENTS ARE REQUIRED TO ATTEND AND
ALL ISE GRADUATE STUDENTS ARE ENCOURAGED TO ATTEND
REFRESHMENTS WILL BE SERVED FOLLOWING THE SEMINAR – IN THE 4TH FLOOR GOTT LOUNGE AREA
Wednesday, February 23, 2011
ISE SEMINAR 2-25-11
INDUSTRIAL AND SYSTEMS ENGINEERING DEPARTMENT
SPRING 2011 SEMINAR SERIES
TITLE: Multiscale Methods For Large-Scale Combinatorial
SPEAKER: Dr. Ilya Safro
DATE / TIME: Friday, February 25, 2011
LOCATION: Room 453 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: In many cases, a big scale gap can be observed between micro- and macroscopic scales of problems because of the difference in mathematical (engineering, social, biological, physical, etc.) models and/or laws at different scales. The main objective of multiscale algorithms is to create a hierarchy of problems, each representing the original problem at different coarse scales with fewer degrees of freedom. We will talk about different strategies of creating these hierarchies for large-scale combinatorial optimization problems: linear ordering, network compression, partitioning, clustering and constrained 2D-layout problem. These strategies are inspired by the classical multigrid frameworks: Geometric Multigrid, Algebraic Multigrid and Full Approximation Scheme. We will present in details a framework for designing linear time Algebraic Multigrid based multiscale algorithm for the linear ordering problems.
Our algorithms were developed for practical purposes and we compared them to many different methods such as: spectral sequencing, decomposition trees, multilevel-based, stochastic searches, genetic algorithms, path relinking, GRASP-based and other (including their combinations). For almost all large-scale instances (about 200 application-based instances), we observed a significant improvement of the results and/or the computational time. Our multiscale methods have proved themselves to be robust both as a first approximation and as more aggressive optimization solvers.
BIOGRAPHY: Ilya Safro studied Mathematics and Computer Science at the Weizmann Institute of Science where he obtained his Ph.D. under the supervision of Achi Brandt and Dorit Ron. Since 2007 Ilya was a postdoctoral fellow at Argonne National Laboratory. Today he is an Argonne Scholar at the Laboratory of Advanced Numerical Simulations at Argonne National Laboratory. His research interests include multiscale methods, network analysis, combinatorial optimization problems and machine learning.
SPRING 2011 SEMINAR SERIES
TITLE: Multiscale Methods For Large-Scale Combinatorial
SPEAKER: Dr. Ilya Safro
DATE / TIME: Friday, February 25, 2011
LOCATION: Room 453 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: In many cases, a big scale gap can be observed between micro- and macroscopic scales of problems because of the difference in mathematical (engineering, social, biological, physical, etc.) models and/or laws at different scales. The main objective of multiscale algorithms is to create a hierarchy of problems, each representing the original problem at different coarse scales with fewer degrees of freedom. We will talk about different strategies of creating these hierarchies for large-scale combinatorial optimization problems: linear ordering, network compression, partitioning, clustering and constrained 2D-layout problem. These strategies are inspired by the classical multigrid frameworks: Geometric Multigrid, Algebraic Multigrid and Full Approximation Scheme. We will present in details a framework for designing linear time Algebraic Multigrid based multiscale algorithm for the linear ordering problems.
Our algorithms were developed for practical purposes and we compared them to many different methods such as: spectral sequencing, decomposition trees, multilevel-based, stochastic searches, genetic algorithms, path relinking, GRASP-based and other (including their combinations). For almost all large-scale instances (about 200 application-based instances), we observed a significant improvement of the results and/or the computational time. Our multiscale methods have proved themselves to be robust both as a first approximation and as more aggressive optimization solvers.
BIOGRAPHY: Ilya Safro studied Mathematics and Computer Science at the Weizmann Institute of Science where he obtained his Ph.D. under the supervision of Achi Brandt and Dorit Ron. Since 2007 Ilya was a postdoctoral fellow at Argonne National Laboratory. Today he is an Argonne Scholar at the Laboratory of Advanced Numerical Simulations at Argonne National Laboratory. His research interests include multiscale methods, network analysis, combinatorial optimization problems and machine learning.
Postdoc position in data analysis/visualization/neuroscience
Job Description: We are seeking two highly motivated postdoctoral fellows in
computational neuroscience to be part of an interdisciplinary research
alliance working to develop mobile brain/body imaging experimental and data
analysis methods in support of a research program in neuroergonomics (‘the
study of the brain and body at work’). An overall goal of the research is to
discover underlying principles describing the relationship of non-invasively
recorded EEG brain dynamics and motivated behavior (recorded by body motion
capture, eye tracking, etc) in interactive, information-rich human-system
operating environments and to apply these principles to support overall
performance of complex system operations.
The ideal candidates will have research experience in neuroscience or
cognitive science with a strong interest in computation, data analysis, or
visualization. The candidates will be based near Baltimore, Maryland at the
Army Research Laboratory (ARL), Aberdeen, MD, where they will collaborate
with a group of Army-funded government and industry researchers in gathering
and analyzing data from successively more complex and realistic experiments.
Successful candidates will be hired by and will interface with academic
research groups at UC San Diego (Scott Makeig, Tzyy-Ping Jung, Roger Levy,
Ken Kreutz-Delgatho, Angela Yu, Falko Kuester), UT San Antonio (Kay Robbins,
Yufei Huang, Nandini Kannan), and/or University of Michigan (Dan Ferris),
with additional ongoing collaborations with groups at National Chiao Tung
University, Taiwan (Chin-Teng Lin), and University of Osnabrück, Germany
(Peter König). The researchers will be encouraged to work closely with
researchers and students at participating universities and to present their
research at conferences and in the open research literature.
Funding Period: Salaries will be competitive. The project is in the first of
five years of funding, with subsequent renewal for an equal period possible.
Transitions to permanent government research positions may be available for
successful candidates.
Minimum Requirements: Ph.D. with research experience in computational
approaches to cognitive psychology or neuroscience. Both beginning and more
senior postdoctoral candidates are encouraged to apply.
Preferred Qualifications: Strong computational skills with experience in
design, analysis, and statistical signal processing/machine learning applied
to data from complex experimental designs. American citizenship preferred.
Instructions to Applicants: Applicants should submit a cover letter and CV,
including the names and contact information of three references. Include in
the cover letter accompanying the application a summary of your research
experience and goals. Please send application materials by e-mail to:
Scott Makeig, Director, Swartz Center for Computational Neuroscience,
Institute for Neural Computation, University of California San Diego;
smakeig@ucsd.edu; http://sccn.ucsd.edu; E-mail subject: ARL Aberdeen
position.
Kay Robbins, Professor, Department of Computer Science, University of Texas
at San Antonio; krobbins@cs.utsa.edu; http://visual.cs.utsa.edu; (210)
458-5543; E-mail subject: ARL Aberdeen position.
The University of California at San Diego and the University of Texas at San
Antonio are Equal Opportunity/Affirmative Action Employers. As part of the
application process, applicants will be invited to complete an online
confidential and voluntary self-disclosure card.
computational neuroscience to be part of an interdisciplinary research
alliance working to develop mobile brain/body imaging experimental and data
analysis methods in support of a research program in neuroergonomics (‘the
study of the brain and body at work’). An overall goal of the research is to
discover underlying principles describing the relationship of non-invasively
recorded EEG brain dynamics and motivated behavior (recorded by body motion
capture, eye tracking, etc) in interactive, information-rich human-system
operating environments and to apply these principles to support overall
performance of complex system operations.
The ideal candidates will have research experience in neuroscience or
cognitive science with a strong interest in computation, data analysis, or
visualization. The candidates will be based near Baltimore, Maryland at the
Army Research Laboratory (ARL), Aberdeen, MD, where they will collaborate
with a group of Army-funded government and industry researchers in gathering
and analyzing data from successively more complex and realistic experiments.
Successful candidates will be hired by and will interface with academic
research groups at UC San Diego (Scott Makeig, Tzyy-Ping Jung, Roger Levy,
Ken Kreutz-Delgatho, Angela Yu, Falko Kuester), UT San Antonio (Kay Robbins,
Yufei Huang, Nandini Kannan), and/or University of Michigan (Dan Ferris),
with additional ongoing collaborations with groups at National Chiao Tung
University, Taiwan (Chin-Teng Lin), and University of Osnabrück, Germany
(Peter König). The researchers will be encouraged to work closely with
researchers and students at participating universities and to present their
research at conferences and in the open research literature.
Funding Period: Salaries will be competitive. The project is in the first of
five years of funding, with subsequent renewal for an equal period possible.
Transitions to permanent government research positions may be available for
successful candidates.
Minimum Requirements: Ph.D. with research experience in computational
approaches to cognitive psychology or neuroscience. Both beginning and more
senior postdoctoral candidates are encouraged to apply.
Preferred Qualifications: Strong computational skills with experience in
design, analysis, and statistical signal processing/machine learning applied
to data from complex experimental designs. American citizenship preferred.
Instructions to Applicants: Applicants should submit a cover letter and CV,
including the names and contact information of three references. Include in
the cover letter accompanying the application a summary of your research
experience and goals. Please send application materials by e-mail to:
Scott Makeig, Director, Swartz Center for Computational Neuroscience,
Institute for Neural Computation, University of California San Diego;
smakeig@ucsd.edu; http://sccn.ucsd.edu; E-mail subject: ARL Aberdeen
position.
Kay Robbins, Professor, Department of Computer Science, University of Texas
at San Antonio; krobbins@cs.utsa.edu; http://visual.cs.utsa.edu; (210)
458-5543; E-mail subject: ARL Aberdeen position.
The University of California at San Diego and the University of Texas at San
Antonio are Equal Opportunity/Affirmative Action Employers. As part of the
application process, applicants will be invited to complete an online
confidential and voluntary self-disclosure card.
Tuesday, February 22, 2011
Emacs Quick Reference
Key Bindings
- Lines
-
C-a beginning-of-line C-e end-of-line C-n next-line C-p previous-line C-k kill-line C-o open-line
- Words
-
ESC f forward-word ESC b backward-word ESC d kill-word ESC DEL backward-kill-word
- Characters
-
C-f forward-char C-b backward-char C-d delete-char DEL delete-backward-char C-q quoted-insert C-t transpose-chars
- Regions
-
C-@ set-mark-command C-w kill-region (between cursor and mark) C-y yank (i.e., insert text last killed)
- Screen control
-
C-l recenter C-v scroll-up (forward) ESC-v scroll-down (backward) ESC < beginning-of-buffer ESC > end-of-buffer
- Search
-
C-s isearch-forward C-r isearch-backward
- Files
-
C-x C-f find-file C-x C-r find-file-read-only C-x C-w write-file
- Windows
-
C-x 1 delete-other-windows C-x 2 split-window-vertically C-x 4 f find-file-other-window C-x o other-window
- Command execution
-
ESC ! shell-command ESC x compile compile ("make -k" is default) C-x ` next-error (used after "compile" to find/edit errors) - Miscellaneous
-
C-x C-c save-buffers-kill-emacs C-u universal-argument C-x C-z suspend-emacs (resume by typing "fg" to unix) - Help!
-
C-g keyboard-quit C-h help-command C-h t help-with-tutorial C-h b describe-bindings (complete list of emacs commands)
Commands for Compiling
ESC-x compile- runs the compiler, linker, etc.
make -k. If you're not using make -k erase the minibuffer line (using DEL, for example) and type in the compiler command of your choice, e.g., lcc -g -A encode.c. This command is remembered for subsequent executions of ESC-x compile. When you type RETURN, if there are unsaved buffers, you will be given the opportunity to save each one. The screen then splits into two windows, and the output from the compilation command appears in one of the two windows. If there are parse errors, use the following command.C-x `- finds the locations of errors.
ESC-x compile command again.Commands for Debugging
ESC-x gdb- runs gdb, the GNU interactive debugger.
-o option to name the output file, the name of your executable file is a.out. When you type in the file name followed by RETURN, the screen splits into two windows (or remain split if it is split already). One window is used for interactive input and output to gdb. The other will eventually display your program files for you to examine and edit. Sometimes the screen doesn't divide immediately after ESC-x gdb, but gdb takes over the whole window where it was executed from; the screen divides the first time you run the program and it stops because of a breakpoint or an error caught by the debugger. So, if the window doesn't split and you want to follow the behaviour of the running program, just type break main before you run it the first time. When execution reaches main, the window splits as described above, an arrow points to the current position in the code, which is the first line of main.The cursor initially is placed after the
gdb prompt (gdb). Whenever you want to issue a command to gdb, position the cursor at the end of the buffer, i.e., after the (gdb), and type the command as usual. The command ESC-> gets you to the end of the buffer. To examine previous input or output to gdb, use the usual emacs commands to move around the buffer.Whenever your program, which was running under
gdb, stops because of a breakpoint an interrupt, etc., the source code associated with the current locus of execution is displayed automatically in the other window. A marker, =>, points to the specific line. If you use the frame command to change frames, the source for the new frame is displayed and the marker is placed accordingly.When you find an error, you may change the source code and save the file. However, before recompiling, give
gdb the command kill to cancel your running program. Otherwise, when the compiler runs the linker to link your program you'll get the error `text file busy' and a new executable file will not be written.After recompiling a program, you should reload the symbol table and the executable, otherwise you'll be running the previous program. To do so, execute
(gdb) exec-file program-name (gdb) symbol-file program-nameThe
symbol-file command will request confirmation before reloading the symbol table; just answer yes.ESC-x gdb-break- sets a
gdbbreakpoint at the source line on which the cursor appears.
Commands for Controlling Windows
C-x 1- reformats screen into one window, retaining only the window in which the cursor appears.
C-x 4f- finds a file and displays it in the other window (the window in which the cursor does not appear). If the screen has only one window, split it into two. The
C-x 4fcommand prompts for the file name. C-x o- moves the cursor to another window.
ISE seminar 2-18
INDUSTRIAL AND SYSTEMS ENGINEERING DEPARTMENT
SPRING 2011 SEMINAR SERIES
TITLE: Some Practical Methods for Computing Sparse or Low-Rank Solutions
SPEAKER: Dr. Zhaosong Lu, Assistant Professor of Mathematics, Simon Fraser University
And associate faculty member in Statistics and Actuarial Science.
DATE / TIME: Friday, February 18, 2011
2:30 – 3:30 p.m.
LOCATION: Room 453 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: Nowadays there are numerous emerging applications in science and engineering concerning about sparse or low-rank solution such as compressed sensing, image recovery and dimension reduction. In this talk, we propose some practical methods for computing sparse or low-rank solutions. In particular, we study a novel first-order augmented Lagrangian (AL) method for solving a class of nonsmooth constrained optimization problems which include l1-norm and nuclear-norm regularized problems as special cases. The global convergence of this method is established. We also develop two first-order methods for solving the associated AL subproblems and establish their global and local convergence. In addition, we propose penalty decomposition methods for solving l0-norm and rank minimization problems. Under some suitable assumptions, we show that each accumulation point is a stationary point of an equivalent smooth optimization problem. Finally, we demonstrate the computational performance of these methods by applying them to sparse principal component analysis and sparse logistic regression.
BIOGRAPHY: Zhaosong Lu is an Assistant Professor of Mathematics at Simon Fraser University and an associate faculty member in Statistics and Actuarial Science. He received his PhD in Operations Research from Georgia Institute of Technology in 2005. He was a Zeev Nehari Visiting Assistant Professor of Mathematical Sciences at Carnegie Mellon University during 2005-2006. Dr. Lu's research interests include theory and algorithms for continuous optimization, and applications in data mining, finance, statistics, machine learning, image processing, engineering design, and decision-making under uncertainty. He was a finalist of 2005 INFORMS George Nicholson Best Student Paper Competition for the work on first-order methods for solving large-scale well-structured semi definite programming. He has published papers in major journals of his research areas such as Mathematical Programming, SIAM Journal on Optimization, INFORMS Journal on Computing, SIAM Journal on Matrix Analysis and Applications, Optimization Methods and Software, Journal of the Royal Statistical Society, and ASME Journal of Mechanical Design.
ALL FULL-TIME ISE PHD STUDENTS ARE REQUIRED TO ATTEND AND
ALL ISE GRADUATE STUDENTS ARE ENCOURAGED TO ATTEND
REFRESHMENTS WILL BE SERVED FOLLOWING THE SEMINAR
SPRING 2011 SEMINAR SERIES
TITLE: Some Practical Methods for Computing Sparse or Low-Rank Solutions
SPEAKER: Dr. Zhaosong Lu, Assistant Professor of Mathematics, Simon Fraser University
And associate faculty member in Statistics and Actuarial Science.
DATE / TIME: Friday, February 18, 2011
2:30 – 3:30 p.m.
LOCATION: Room 453 Mohler Lab, 200 W. Packer Avenue
ABSTRACT: Nowadays there are numerous emerging applications in science and engineering concerning about sparse or low-rank solution such as compressed sensing, image recovery and dimension reduction. In this talk, we propose some practical methods for computing sparse or low-rank solutions. In particular, we study a novel first-order augmented Lagrangian (AL) method for solving a class of nonsmooth constrained optimization problems which include l1-norm and nuclear-norm regularized problems as special cases. The global convergence of this method is established. We also develop two first-order methods for solving the associated AL subproblems and establish their global and local convergence. In addition, we propose penalty decomposition methods for solving l0-norm and rank minimization problems. Under some suitable assumptions, we show that each accumulation point is a stationary point of an equivalent smooth optimization problem. Finally, we demonstrate the computational performance of these methods by applying them to sparse principal component analysis and sparse logistic regression.
BIOGRAPHY: Zhaosong Lu is an Assistant Professor of Mathematics at Simon Fraser University and an associate faculty member in Statistics and Actuarial Science. He received his PhD in Operations Research from Georgia Institute of Technology in 2005. He was a Zeev Nehari Visiting Assistant Professor of Mathematical Sciences at Carnegie Mellon University during 2005-2006. Dr. Lu's research interests include theory and algorithms for continuous optimization, and applications in data mining, finance, statistics, machine learning, image processing, engineering design, and decision-making under uncertainty. He was a finalist of 2005 INFORMS George Nicholson Best Student Paper Competition for the work on first-order methods for solving large-scale well-structured semi definite programming. He has published papers in major journals of his research areas such as Mathematical Programming, SIAM Journal on Optimization, INFORMS Journal on Computing, SIAM Journal on Matrix Analysis and Applications, Optimization Methods and Software, Journal of the Royal Statistical Society, and ASME Journal of Mechanical Design.
ALL FULL-TIME ISE PHD STUDENTS ARE REQUIRED TO ATTEND AND
ALL ISE GRADUATE STUDENTS ARE ENCOURAGED TO ATTEND
REFRESHMENTS WILL BE SERVED FOLLOWING THE SEMINAR
Monday, February 14, 2011
Makefile.am and Makefile.in
Makefile.amis theautomakesource file.Makefile.inis generated byautomakefromMakefile.am
Subscribe to:
Posts (Atom)