Skip Navigation
Oklahoma State University

IEM Seminar Series

The School of Industrial Engineering and Management in conjunction with the OSU INFORMS Student Chapter sponsors a Seminar Series in the fall and spring semesters. Various topics are covered by speakers from OSU and other organizations. The seminars are held on Wednesdays from 1-2:30 pm in the Fall semester and on Thursdays from 3:30-5:00 pm in the Spring semester. Seminar topics, location and a short abstract and bio about the presenters are posted below as it becomes available.

 

Date

Speaker

Title

Institution

Jan. 17th

Dr. Ashesh Kumar Sinha

Stochastic Models for Strategic Sourcing in Supply Chains

Schneider International

Jan. 19th

Mr. Hao Yan

Sequential High-Dimensional Data Analysis for Anomaly Detection and System Monitoring

Georgia Institute of Technology

Jan. 24th

Dr. Devashish Das

Data Analytics for Complex Systems: Improving the Science of Health Care Delivery

Mayo Clinic

Jan. 26th

Mr. Juan Borrero

Sequential Max-Min Bilevel Programming with Incomplete Information and Learning

University of Pittsburgh

Mar. 9th

Mr. G. Satish

SMAC Impact-Social, Mobile, Analytics, Cloud, and the Engineer

Connixt

Mar. 23rd

Dr. Suvrajeet Sen

Learning Enabled Optimization: Towards a Fusion of Stochastical Learning and Stochastic Optimization

University of Southern California

Mar. 24th

Dr. Julie Higle*

Modeling and Analysis for Cancer Screening

University of Southern California

Mar. 30th

Dr. Art Chaovalitwongse

Optimization in Medical Analytics: From Data to Knowledge to Decisions

University of Arkansas

Apr. 13th

Dr. James Tien

Internet of Things, Real-Time Decision Making, and Artificial Intelligence

University of Miami

Apr. 20th

Dr. David Morton

Optimizing Prioritized and Nesting Solutions

Northwestern University

*This seminar occurs on Friday at a time to be determined.

 

Industrial Engineering and Management Seminar Series
Sponsored by IEM and OSU INFORMS Student Chapter

Learning Enabled Optimization: Towards a Fusion of Statistical Learning and Stochastic Optimization 

Speaker: Dr. Suvrajeet Sen

Date: Thursday, March 23rd 2017

Time: Seminar - 3:30 to 4:20 p.m Q & A / Social- 4:20 to 5:00 p.m.

Location: Engineering North 450

 

Abstract: Several emerging applications, such as “Analytics of Things” and “Integrative Analytics” call for a fusion of statistical learning (SL) and stochastic optimization (SO). The Learning Enabled Optimization paradigm fuses concepts from these disciplines in a manner which not only enriches both SL and SO, but also provides a framework which supports rapid model updates and optimization, together with a methodology for rapid model-validation, assessment, and selection.  Moreover, in many “big data/big decisions” applications these steps are repetitive, and possible only through a continuous cycle involving data analysis, optimization, and validation. This talk sets forth the foundation for such a framework by introducing several novel concepts such as statistical optimality, hypothesis tests for model-fidelity, generalization error of stochastic optimization, and finally, a non-parametric methodology for model selection.   We illustrate the LEO framework by applying it to an inventory control model in which we use demand data available for ARIMA modeling in the statistical package “R”.  In addition, we also study a production-marketing coordination model based on combining a pedagogical production planning model with an advertising data set intended for sales prediction. This talk only scratches the surface of a new genre of models which we believe will be critical for many applications in the future.  (This talk is joint work with my Ph.D. student Ms. Yunxiao Deng.)

Speaker Bio: Suvrajeet Sen is Professor at the Daniel J. Epstein Department of Industrial and Systems Engineering at the University of Southern California.  Prior to joining USC, he was a Professor at Ohio State University (2006-2012), and University of Arizona (1982-2006). He has also served as the Program Director of OR as well as Service Enterprise Systems at the National Science Foundation. Professor Sen’s research is devoted to many categories of optimization models, and he has published over one hundred papers, with the vast majority of them dealing with models, algorithms and applications of Stochastic Programming problems.  He has served on several editorial boards, including Operations Research as Area Editor for Optimization and as Associate Editor for INFORMS Journal on ComputingJournal of Telecommunications SystemsMathematical Programming B, and Operations Research.   He also serves as an Advisory Editor for several newer journals.  Professor Sen was instrumental in founding the INFORMS Optimization Society in 1995, and recently served as its Chair (2015-16).  Except for his years at NSF, he has received continuous extramural research funding from NSF and other basic research agencies, totaling over nine million dollars as PI over his career.  His group’s contributions were recognized by the INFORMS Computing Society for “seminal work” on Stochastic Mixed-Integer Programming.  Professor Sen is a Fellow of INFORMS.

 

Modeling and Analysis for Cancer Screening

Speaker: Julie Higle

Chair of the Daniel J. Epstein Department of Industrial and Systems Engineering

University of Southern California, Los Angeles, CA

Date: Friday, March 24th  2017

Time: TBA

Location: Engineering North 450

 

Abstract: Cancer screening strategies facilitate early detection of cancer in a systematic fashion.  Early detection can lead to improved treatment prospects, increased survival rates, improved quality of life for survivors, and reduced treatment costs.  Increasingly often, model-based analyses of screening and treatment strategies are used to inform health policy and its implementation. They permit an exploration of a broader range of strategies than might be tested in clinical studies, including those that are configured hypothetically. 

A central component of a model-based analysis is the natural history model, which represents the evolution of disease in the absence of medical intervention.   The development of a natural history model is an intricate process, requiring significant navigation around issues involving “data” and “model parameters”.  The construction of the model involves various data sources, and modeling techniques are necessary to estimate data that are not available through clinical studies.

This seminar will discuss simulation-based analysis of the relative performance of various strategies for cancer screening, with particular emphasis on cervical cancer.  The construction of the model involves various data sources, and modeling techniques are used to estimate data that are not available through clinical studies. A healthy discussion of the pros and cons of various modeling decisions will undoubtedly be included.   

Speaker Bio: Julie Higle serves as Professor and Chair of the Daniel J. Epstein Department of Industrial and Systems Engineering at the University of Southern California.  Prior to joining USC, she has served as a member of the faculty of Systems and Industrial Engineering at the University of Arizona, and as the chair of the Department of Integrated Systems Engineering at The Ohio State University.    Her research interests are primarily in the development of models and solutions methods for decision making under uncertainty, with a healthy emphasis on medical decision making.