Category: ICC

Dr. Qun Li to Present Lecture April 23, 3 pm


The Department of Computer Science will present a lecture by Dr. Qun Li on Friday, April 23, 2021, at 3:00 p.m. Dr. Li is a professor in the computer science department at William and Mary university. The title of his lecture is, “Byzantine Fault Tolerant Distributed Machine Learning.”

Join the virtual lecture here.

Lecture Title

Byzantine Fault Tolerant Distributed Machine Learning

Lecture Abstract

Training a deep learning network requires a large amount of data and a lot of computational resources. As a result, more and more deep neural network training implementations in industry have been distributed on many machines. They can also preserve the privacy of the data collected and stored locally, as in Federated Deep Learning.

It is possible for an adversary to launch Byzantine attacks to a distributed or federated deep neural network training. That is, some participating machines may behave arbitrarily or maliciously to deflect the training process. In this talk, I will discuss our recent results on how to make distributed and federated neural network training resilient to Byzantine attacks. I will first show how to defend against Byzantine attacks in a distributed stochastic gradient descent (SGD) algorithm, which is the core of distributed neural network training. Then I will show how we can defend against Byzantine attacks in Federated Learning, which is quite different from distributed training.


Article by Sidike Paheding in Elsevier’s Remote Sensing of Environment


An article by Dr. Sidike Paheding, Applied Computing, has been accepted for publication in the Elsevier journal, Remote Sensing of Environment, a top journal with an impact factor of 9.085. The journal is ranked #1 in the field of remote sensing, according to Google Scholar.

The paper, “Estimation of root zone soil moisture from ground and remotely sensed soil information with multisensor data fusion and automated machine learning,” will be published in Volume 260, July 2021 of the journal. Read and download the article here.

Highlights

  • A machine learning approach to estimation of root zone soil moisture is introduced.
  • Remotely sensed optical reflectance is fused with physical soil properties.
  • The machine learning models well capture in situ measured root zone soil moisture.
  • Model estimates improve when measured near-surface soil moisture is used as input.

Paheding’s co-authors are:

  • Ebrahim Babaeian, Assistant Research Professor, Environmental Science, University of Arizona, Tucson
  • Vijay K. Devabhaktuni, Professor of Electrical Engineering, Department Chair, Purdue University Northwest, Hammond, IN
  • Nahian Siddique, Graduate Student, Purdue University Northwest
  • Markus Tuller, Professor, Environmental Science, University of Arizona

Abstract

Root zone soil moisture (RZSM) estimation and monitoring based on high spatial resolution remote sensing information such as obtained with an Unmanned Aerial System (UAS) is of significant interest for field-scale precision irrigation management, particularly in water-limited regions of the world. To date, there is no accurate and widely accepted model that relies on UAS optical surface reflectance observations for RZSM estimation at high spatial resolution. This study is aimed at the development of a new approach for RZSM estimation based on the fusion of high spatial resolution optical reflectance UAS observations with physical and hydraulic soil information integrated into Automated Machine Learning (AutoML). The H2O AutoML platform includes a number of advanced machine learning algorithms that efficiently perform feature selection and automatically identify complex relationships between inputs and outputs. Twelve models combining UAS optical observations with various soil properties were developed in a hierarchical manner and fed into AutoML to estimate surface, near-surface, and root zone soil moisture. The addition of independently measured surface and near-surface soil moisture information to the hierarchical models to improve RZSM estimation was investigated. The accuracy of soil moisture estimates was evaluated based on a comparison with Time Domain Reflectometry (TDR) sensors that were deployed to monitor surface, near-surface and root zone soil moisture dynamics. The obtained results indicate that the consideration of physical and hydraulic soil properties together with UAS optical observations improves soil moisture estimation, especially for the root zone with a RMSE of about 0.04 cm3 cm−3. Accurate RZSM estimates were obtained when measured surface and near-surface soil moisture data was added to the hierarchical models, yielding RMSE values below 0.02 cm3 cm−3 and R and NSE values above 0.90. The generated high spatial resolution RZSM maps clearly capture the spatial variability of soil moisture at the field scale. The presented framework can aid farm scale precision irrigation management via improving the crop water use efficiency and reducing the risk of groundwater contamination.


Remote Sensing of Environment (RSE) serves the Earth observation community with the publication of results on the theory, science, applications, and technology of remote sensing studies. Thoroughly interdisciplinary, RSE publishes on terrestrial, oceanic and atmospheric sensing. The emphasis of the journal is on biophysical and quantitative approaches to remote sensing at local to global scales.


Get to Know Dr. Sangyoon Han, Biomedical Engineering


Dr. Sangyoon Han is an assistant professor in Michigan Tech’s Biomedical Engineering department, and an affiliated assistant professor in the Mechanical Engineering-Engineering Mechanics department. He is also advisor to the Korean Students Association. He has been with Michigan Tech since 2017.

Han recently joined the Institute of Computing and Cybersystems and its Data Sciences research group. His primary research interests are in mechanobiology, cell migration, and image data modeling. His research goals include applying computer vision to microscopic images to capture meaningful information, and he’s looking for collaborators.

“Anyone with a good machine learning background is encouraged to contact me to discuss potential research,” he says. “Also, students who learned assignment problems or particle tracking are encouraged to contact me to discuss potential tracking-related projects.”

Teaching and Mentoring

Han’s teaching interests include computer vision for microscopic images, fluid mechanics, cell biomechanics and mechanobiology, and soft tissue mechanics. This academic year, he instructed Computer Vision for Microscopic Images in the Fall semester, and Fluid Mechanics this Spring.

Han enjoys teaching and interacting with students, “and feel their energy, too.” He says he makes a deliberate effort in his classes to pause from time to time so that his students can ask questions.

Han advises two Biomedical Engineering Ph.D. students, Nikhil Mittal and Mohanish Chandurkar.

“Nik is working on finding myosin-independent mechanosensitivity mechanism for stiffness sensing, and Mohanish works on the project finding mechano-transmission for fluid shear stress sensing by endothelial cells,” he says.

Research Aspirations

Han’s Mechanobiology Lab is interested in finding fundamental mechanisms governing mechanotransduction, and how cells sense mechanical forces and convert them into biochemical signals.

“We image cells and associated forces using high-resolution live imaging, which we analyze to obtain statistically meaningful quantity of data,” Han explains. “We apply force-measuring and molecular-imaging/analysis technologies for stiffness sensing, shear flow sensing, adhesion assembly, and cancer mechanobiology.”

Han is working to gain a thorough understanding of the mechano-chemical interaction between cancer cells and their microenvironment, and develop a an effective mechano-therapeutic strategy to stop the progression of cancer, and breast cancer in particular. Ultimately, he wants to apply that knowledge to cancer mechanobiology

Han is principal investigator of a three-year NIH/NIGMS research project, “Nascent Adhesion-Based Mechano-transmission for Extracellular Matrix Stiffness Sensing.” The research aims to determine whether newly-born adhesions can sense tissue stiffness through the accurate measurement of the mechanical force and molecular recruitment of early adhesion proteins.

Some Background

In 2012, Han received his Ph.D. in Mechanical Engineering from the University of Washington in the areas of cell mechanics, multiphysics modeling, and bioMEMS.

For his postdoctoral training, he joined the Computational Cell Biology lab led by Dr. Gaudenz Danuser in the Cell Biology department of Harvard Medical School. In 2014, he joined the UT Southwestern (University of Texas) Department of Cell Biology and Bioinformatics. Han received his B.S and M.S. degrees in mechanical engineering at Seoul National University, Korea, in 2002 and 2004, respectively.

Han holds several patents and in 2015, he developed an open-source TFM (Traction Force Microscopy) Package, which is shared via his lab’s website: hanlab.biomed.mtu.edu/software.

Beyond Research and Teaching

Han loves science and discovering something new in his research investigations. Beyond his work as a professor and scientist, he describes himself as a husband to Sunny, and a dad to his son, Caleb.

“I am just a normal Korean who likes singing and dancing,” he says. “Unfortunately, my voice is still recovering from surgery, but I hope to get back to it soon. I also like to listen to all kinds of music, including hip-hop, classics, and pop.”

He appreciates a good sense of humor, but he says that being humorous in American English is something he continues to learn.

Han says he tries to be “normal” and not too nerd-like when he’s not pursuing his research, but “there are times when I am making my own hypothesis about some phenomena I observe in my daily life.”

Han enjoys life at Michigan Tech and in the Cooper Country. He likes getting to know his energetic students and he finds Michigan Tech faculty members very strong and collegial. He also enjoys the snow, hockey, and the mountains.

“I really like the snow here. I am already sad that the weather is becoming too mild!” he confirms. “It’s also a safe environment to raise kids, which is a big plus.”

And he likes his academic department. “Everyone is so nice in the Biomedical Engineering program, they have been so welcoming and appreciative my research,” Han says. “It’s a family-like environment.”


Active Research

1R15GM135806-01 (09/16/2019 – 08/31/2022)

Funding Agency: NIH/NIGMS

Nascent Adhesion-Based Mechano-transmission for Extracellular Matrix Stiffness Sensing

Project Goals: To determine whether newly-born adhesions can sense tissue stiffness by accurate measurement of mechanical force and of molecular recruitment of early adhesion proteins using traction force microscopy and computer vision techniques.
Role: Principal Investigator


Additional Information

The Mechanobiology Lab studies mechanobiology, particularly how adherent cells can sense and respond to mechanical stiffness of the extracellular matrix. To investigate this, the lab has established experimental and computational frameworks for force measurement and adhesion dynamics quantification. Researchers apply these frameworks, with cutting edge computer vision technique, on live-cell microscope images to investigate the fundamental mechanism underlying mechanosensation in normal cells, and the biomechanical signature of the diseased cells whose signaling has gone awry.

The Institute of Computing and Cybersystems (ICC) creates and supports an arena in which faculty and students work collaboratively across organizational boundaries in an environment that mirrors contemporary technological innovation. The ICC’s 60+ members, in six research centers, represent more than 20 academic disciplines at Michigan Tech. https://www.mtu.edu/icc/

The ICC Center for Data Sciences (DataS) focuses on the research of data sciences education, algorithms, mathematics, and applications. https://www.mtu.edu/icc/centers/data-sciences/

The National Institutes of Health (NIH), a part of the U.S. Department of Health and Human Services, is the nation’s medical research agency — making important discoveries that improve health and save lives. https://www.nih.gov/

The National Institute of General Medical Sciences (NIGMS) supports basic research that increases understanding of biological processes and lays the foundation for advances in disease diagnosis, treatment, and prevention. https://www.nigms.nih.gov/


Recent Publications

  • Han, S. J.; Azarova, E. V.; Whitewood, A. J.; Bachir, A.; Guttierrez, E.; Groisman, A.; Horwitz, A. R.; Goult, B. T.; Dean, K. M.; Danuser, G. Pre-Complexation of Talin and Vinculin without Tension Is Required for Efficient Nascent Adhesion Maturation. eLife 2021, 10, e66151. https://doi.org/10.7554/eLife.66151.
  • Schäfer, C., Ju, Y., Tak, Y., Han, S.J., Tan, E., Shay, J.W., Danuser, G., Holmqvist, M., Bubley, G. (2020) TRA-1-60-positive cells found in the peripheral blood of prostate cancer patients correlate with metastatic disease. Heliyon 6(1), e03263.
  • Isogai, T., Dean, K.M., Roudot, P., Shao, Q., Cillay, J.D., Welf, E.S., Driscoll, M.K., Royer, S.P., Mittal, N., Chang, B., Han, S.J., Fiolka, R., Danuser, G., Direct Arp2/3-vinculin binding is essential for cell spreading, but only on compliant substrates and in 3D, BioRxiv, 2019
  • Mohan, A.S., Dean, K.M., Isogai, T., Kasitinon, S.Y., Murali, V.S., Roudot, P., Groisman, A., Reed, D.K., Welf, E.S., Han, S.J., Noh, J., and Danuser, G. (2019). Enhanced Dendritic Actin Network Formation in Extended Lamellipodia Drives Proliferation in Growth-Challenged Rac1P29S Melanoma Cells. Developmental Cell, 49(3), pp.444-460.
  • Manifacier I., Milan, J., Beussman, K., Han, S.J., Sniadecki, N.J., About, I (2019) The consequence of large-scale rigidity on actin network tension. In press. Comp Meth Biomech Biomed Eng, 2019 Oct;22(13):1073-1082.
  • Costigliola, N., Ding, L., Burckhardt, C.J., Han, S.J., Gutierrez, E., Mota, A., Groisman, A., Mitchison, T.J., and Danuser, G. (2017) Vimentin directs traction stress. PNAS2017 114 (20) 5195-5200.
  • Han, S.J., Rodriguez M.L., Al-Rekabi, Z., Sniadecki, N.J. (2016) Spatial and Temporal Coordination of Traction Forces in One-Dimensional Cell Migration, Cell Adhesion & Migration. 10(5): 529-539.
  • Oudin, M.J., Barbier, L., Schäfer, C, Kosciuk, T., Miller, M.A., Han, S.J., Jonas, O., Lauffenburger, D.A., Gertler, F.B. (2016) Mena confers resistance to Paclitaxel in triple-negative breast cancer. Mol Cancer Ther.DOI: 10.1158/1535-7163. MCT-16-0413. 
  • Milan,J., Manifacier, I., Beussman, K.M., Han, S.J., Sniadecki, N.J., About, I., Chabrand, P. (2016) In silico CDM model sheds light on force transmission in cell from focal adhesions to nucleus. J Biomechanics. 49(13):2625-2634. 
  • Lomakin. A.J., Lee, K.C., Han, S.J., Bui, A., Davidson, M., Mogilner, A., Danuser G. (2015) Competition for molecular resources among two structurally distinct actin networks defines a bistable switch for cell polarization, Nature Cell Biology. 17, 1435–1445
  • Han, S.J., Oak, Y., Groisman, A., Danuser, G. (2015) Traction Microscopy to Identify Force Modulation in Sub-resolution Adhesions, Nature Methods. 12(7): 653–656

1010 with … Dr. Alex Sergeyev, Applied Computing


Are you a high school student, current undergraduate student, or a recent BS graduate? Are you are interested in robotics, automation, and controls?

“If you’d like to learn more about the Mechatronics and the BS and MS programs at Michigan Tech, please join this 1010 conversation,” Professor Alex Sergeyev urges.

You are invited to spend one-zero-one-zero—that is, ten—minutes with Dr. Aleksandr Sergeyev on Thursday, April 15, from 4:30 to 4:40 p.m. EST.

Join the Zoom meeting here.


Dr. Sergeyev is a professor in the Applied Computing department and director of the Mechatronics graduate program. He also directs the FANUC Certified Industrial Robotics Training Center at Michigan Tech.

Dr. Sergeyev will discuss his research, the Applied Computing department, and the Mechatronics BS and MS programs. He will answer questions following his presentation.

Michigan Tech is a pioneer in Mechatronics education, having introduced a graduate degree program in 20xx, and a bachelor’s program in Fall 2019.

“Mechatronics is an industry buzzword synonymous with robotics, controls, automation, and electromechanical engineering,” Sergeyev says.

In his presentation, he will discuss Mechatronics in general, explain what the degree has to offer, job opportunities in Mechatronics, and some of the research he is conducting in this field.

In Spring 2021, a Mechatronics Playground was opened on campus. The hands-on learning lab and industry-grade equipment was funded by alumnus Mark Gauthier of Donald Engineering, Grand Rapids, MI, and other major companies.

A common degree in Europe, China, Japan, Russia, and India, advanced study in Mechatronics is an underdeveloped academic discipline in the United States, even though the industrial demand for these professionals is enormous, and continues to grow.

Sergeyev’s areas of expertise are in electrical and computer engineering, physics, and adaptive optics, and his professional interests include robotics. He is principal investigator for research grants totaling more that $1 million. He received both his MS and PhD degrees at Michigan Tech, in physics and electrical and computer engineering, respectively.

We look forward to spending 1010 minutes with you!


Sangyoon Han Publishes Paper in eLife

eLife, a prestigious journal in cell biology, has published a paper co-written by Sangyoon Han, “Pre-complexation of talin and vinculin without tension is required for efficient nascent adhesion maturation.”

Dr. Han is an assistant professor in the Biomedical Engineering department, and a member of the Data Sciences research group of the Institute of Computing and Cybersystems (ICC).

View the paper here.

eLife is a non-profit organization created by funders and led by researchers. Their mission is to accelerate discovery by operating a platform for research communication that encourages and recognizes the most responsible behaviors.


Sidike Paheding, Applied Computing, Publishes Paper in IEEE Access

A paper co-authored by Sidike Paheding, Applied Computing, has been published in the journal, IEEE Access. “Trends in Deep Learning for Medical Hyperspectral Image Analysis,” was available for early access on March 24, 2021.

The paper discusses the implementation of deep learning for medical hyperspectral imaging.

Co-authors of the paper are Uzair Khan, Colin Elkin, and Vijay Devabhaktuni, all with the Department of Electrical and Computer Engineering, Purdue University Northwest.

Abstract

Deep learning algorithms have seen acute growth of interest in their applications throughout several fields of interest in the last decade, with medical hyperspectral imaging being a particularly promising domain. So far, to the best of our knowledge, there is no review paper that discusses the implementation of deep learning for medical hyperspectral imaging, which is what this work aims to accomplish by examining publications that currently utilize deep learning to perform effective analysis of medical hyperspectral imagery.

This paper discusses deep learning concepts that are relevant and applicable to medical hyperspectral imaging analysis, several of which have been implemented since the boom in deep learning. This will comprise of reviewing the use of deep learning for classification, segmentation, and detection in order to investigate the analysis of medical hyperspectral imaging. Lastly, we discuss the current and future challenges pertaining to this discipline and the possible efforts to overcome such trials.

DOI: 10.1109/ACCESS.2021.3068392

IEEE Access is a multidisciplinary, applications-oriented, all-electronic archival journal that continuously presents the results of original research or development across all of IEEE’s fields of interest. Supported by article processing charges, its hallmarks are a rapid peer review and publication process with open access to all readers.


Our Stories: Dr. Robert Pastel, Assoc. Prof., Computer Science

This is part of a series of short introductions about College students, faculty, and staff that we would like to include in the Weekly Download. Would you like to be featured? Send a photo and some background info about yourself to computing@mtu.edu.

Dr. Robert Pastel, Associate Professor of Computer Science

  • Advisor to Humane Interface Design Enterprise (HIDE)
  • Has been teaching at Michigan Tech for about 20 years, and teaching for 30 years.
  • Researcher with the Human-Centered Computing group of the Institute of Computing and Cybersystems (ICC)

Education

  • PhD, University of New Mexico, Physics
  • MS, Computer Science, Michigan Tech

Faculty Profile


Classes Dr. Pastel teaches: 
o    CS5760 – Human-Computer Interaction – Usability Evaluation and Testing 
o    CS4791 and CS4792 – Senior Design
o    ENT1960 – ENT5960 – Humane Interface Design Enterprise

The “coolest” class you teach, and why: All my classes are “cool” because they all involve making applications that will be used by people. The “coolest” class is CS4760 – User Interface – Design and Implementation where students work with scientists across the world to make citizen science applications.

The importance of your class topics to the overall understanding of Computing and your discipline: In all my classes, students learn to design and implement usable applications for people.

Your teaching philosophy: My teaching philosophy is that students learn best by experience and working with others. Consequently students work in teams on project for clients. 

Research projects in which students are assisting: 

  • StreamCLIMES – Large collaborative project studying bio diversity of intermittent streams. I’m responsible for developing a web applications monitoring the stream.
  • FloodAware – Large collaborative project recording and modelling flooding in urban areas. I’m responsible for developing the citizen science effort.
  • KeTT – Keweenaw Time Traveler – Historical geospatial information citizen science website for user to record region’s history and explore the maps and stories. 

Interests beyond teaching and research: The outdoors: skiing, biking and hiking. Every summer, he takes a one-month backpacking trip. 


Visit with Dean Livesay … In Person!

College of Computing Dean Dennis Livesay holds open drop-in office hours every Friday from 3:00 to 4:00 p.m., when classes are in session.

And starting Friday, March 19, you can meet with Dean Livesay in person!

Drop-in office hours are now both virtual and in-person. Stop by Rekhi Hall Room 217, or link to the virtual meeting here: https://michigantech.zoom.us/j/83846079187.

All faculty, staff, and students who wish to chat with Dr. Livesay are invited to “stop in” to this weekly meeting. Appointments are not needed.


Summer Youth Programs (SYP): Topics in Computing

With extensive safety planning and health precautions underway, Michigan Tech Summer Youth Programs plans to offer in-person programs for summer 2021. Programs run weekly from June 21-August 7, 2021.

Registration is now open for 2021 Summer Youth Programs. Many classes are already full, but there are plenty more to choose from

Interested in computing-related classes? Below are SYP programs of particular interest.

Explore the SYP website and see all SYP classes here.

Computing Programs
Class NumberTitleAdditional Cost RequiredSeats AvailableGradesWeek
51400App and Web Development: Designing for Humans129 – 11July 18 – July 24
51890Coding for the Internet of ThingsSee Course Details129 – 11July 11 – July 17
51678Coding for the Internet of ThingsSee Course Details129 – 11June 20 – June 26
52422Introduction to Computational Physics159 – 11June 20 – June 26
51204Introduction to Video Game Programming126 – 8June 27 – July 03
51541Video Game Programming79 – 11July 18 – July 24
Engineering Programs
Class No.Class TitleAdd’l CostsSeats Avail.Grade LevelDates of Class
52409AI & Machine LearningNone89-11July 18 – July 24
52199The Gaming Industry Wants You!None69-11June 27 – July 3
52410Intro to the Perfect MachineNone76-8July 18 – July 24
52412The Perfect MachineNone209-11July 11 – July 17
51909Electrical and Computer EngineeringSee Course Details79-11June 27 – July 3
52092Electrical and Computer EngineeringSee Course Details119-11June 20 – June 26
51190Electrical and Computer EngineeringSee Course Details59-11July 11 – July 17
Scholarship Programs
51435Women in Computer Science (WICS)None179-11June 27 – July 3
Science and Technology Programs
52199The Gaming Industry Wants You!None69-11June 27 – July 3


CS Lecture: Kelly Steelman, CLS, March 19, 3 pm

The Department of Computer Science will present a lecture by Dr. Kelly Steelman, Cognitive and Learning Sciences, on Friday, March 19, 2021, at 3:00 p.m.

The title of the lecture is, “Keeping Up with Tech.”

Join the virtual lecture here.

Steelman is interim department chair and associate professor in the Department of Cognitive and Learning Sciences. Her research interests include basic and applied attention, models of attention, human performance in aviation, display design, tech adoption, and technology training.

Lecture Title

“Keeping Up with Tech”

Lecture Abstract

COVID has revealed much in the past year, including our dependence on technology and the challenges that many of us experience trying to keep up with it. Dr. Kelly Steelman has spent the past 15 years studying human attention and applying it to support the introduction of new technologies in contexts ranging from aviation to education.

In her presentation, Steelman will provide an overview of her research, using examples from Next Gen Aviation and the BASIC Digital Literacy Training Program to illustrate how understanding human attention can help us predict the consequences of introducing new technology, improve the design of technology, and support training to help people keep up with the rapid pace of technological change.