Current Series

[View Past Series]

Fri Sep 28

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
If a tree falls in the woods and nobody hears, does it make a sound? (Or, making sure your code runs anywhere.)
Tyler Whitehouse, Gigantum

Most researchers don’t have the time or the skill to apply the kinds of software development best practices needed to make their computational work transparent, reproducible, and easy to use. Despite repeated calls by publishers and funders to include usable and understandable code with publications, it is still too much work to be done with the resources allotted. Such problems are not unique to academia, and the penetration of machine learning and data science in industry has brought these issues to the attention of commercial enterprise as well. Reproducibility is a problem for everyone.

This talk will demonstrate an open source data science platform that automates the issues around transparency, reproducibility, and ease of use for work done in open source languages and frameworks like Python and R. It will show how easy it is to set up computational and data science environments of varying complexity that can be shared with anybody in the world with no extra labor or set up.

Bio
Tyler Whitehouse did an undergraduate degree in math at UC Santa Cruz and a Ph.D. at the University of Minnesota. He graduated in 2009 after working with Professor Gilad Lerman on problems dealing with the rectifiability of sets and measures in Hilbert spaces, then going on to Vanderbilt University as a postdoc for 3 years. From 2012 to 2017 he worked as a data scientist and consultant in the Washington DC area. Currently, he is the president of a data science software startup in the DC area.

Fri Oct 12

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
Lecture
Jennifer Schumacher, 3M
Fri Nov 02

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 409
Lecture
Joao Montero, Medtronic
Fri Nov 16

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
Advanced Dynamics Systems Analysis and Synthesis for Future Storage Challenges
Raye Sosseh, Seagate Technology

Seagate is the global leader in data storage solutions, developing amazing products that enable people and businesses around the world to create, share and preserve their most critical memories and business data. Over the years the amount of information stored has grown from megabytes all the way to geopbytes, confirming the need to successfully store and access huge amounts of data. As demand for storage technology grows the need for greater efficiency and more advanced capabilities continues to evolve. Today data storage is more than just archiving; it’s about providing ways to analyze information, understand patterns and behavior, to re-live experiences and memories. It’s about harnessing stored information for growth and innovation. Seagate is building on its heritage of storage leadership to solve the challenge of getting more out of the living information that’s produced everyday. What began with one storage innovation has morphed into many systems and solutions becoming faster, more reliable and expansive. These considerations will be explored in this lecture on the application of Advanced Dynamics System Analysis & Synthesis to provide Solutions for future Storage Challenges.

Bio
Dr. Raye A. Sosseh is a managing Principal engineer and leads the Advanced Dynamics Group in Seagate’s Advanced Product Development department. His background is in dynamics and controls of embedded systems with particular applications to Hard Disk Drives. Dr. Sosseh holds a Ph.D. in Mechanical Engineering and a Masters in Electrical Engineering from the Georgia Institute of Technology. He joined Seagate in 2001 and has contributed to the staging and development of numerous products in Seagate’s Mission Critical and Cloud solutions swimlanes. Dr. Sosseh has over 10 patents assigned and pending patents. Raye and his family enjoy the lakes and restaurants in the Minneapolis area.

Fri Nov 30

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
Recap of MinneMUDAC: Challenges and Improvements
-, -

Graduate-level participants of 2018 MinneMUDAC data challenge will give an updated version of their original presentation, supplemented with suggestions for potential improvements to answering the challenge question, which was predicting the voter turnout for the 2018 midterm elections in Minnesota. After the talks, all participants will break into discussions regarding the presentations and the questions that arise from them.

Speakers: Several teams that competed in MinneMUDAC

Fri Dec 07

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
When Seeing is not Believing: New Forensics Algorithms to Detect Image Manipulations
Michael Albright, Honeywell

Modern image editing software has made it possible to alter images in ways that can dramatically change the image content, yet the images may appear authentic to humans. While there are countless beneficial applications of photo editing, image manipulations can also be used in harmful ways – e.g., altered images may be published to cause reputational harm, sway public opinions, influence elections, etc. Furthermore, with the growing popularity of social media and online sharing platforms, it is increasingly easy for altered media to go viral. What’s more, recent breakthroughs in artificial intelligence (AI) are making it dramatically easier to produce altered images, and even altered videos, that appear photo realistic. Hence, there is growing interest in developing new forensic methods that can detect manipulations in images and video. In this talk, I will give an overview of my team’s work on the DARPA-funded Media Forensics (MediFor) Project to develop novel machine-learning algorithms for automated detection and localization of media manipulations.

Michael Albright is a senior data scientist in the Data Science and Video Analytics group in Honeywell Labs in Golden Valley, Minnesota. Since joining Honeywell in 2015, he has invented new technologies that solve challenging problems for internal Honeywell businesses and external customers. His work has involved applying machine learning, optimization, and other applied math and computer science techniques to a variety of problems in domains ranging from Internet of Things (IoT) systems to computer vision applications. Michael earned a Ph.D. in theoretical physics from the University of Minnesota in 2015 and has previous industry experience at Cray, Inc.

Fri Jan 25

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
Lecture
Eric Lind, Metro Transit
Fri Feb 08

IMA MCIM Industrial Problems Seminar

1:25pm - Lind 305
Lecture
Karyn Sutton, The Institute for Disease Modeling