by Mechanical Engineering – Engineering Mechanics
The next virtual Graduate Seminar Speaker will be held at 4 p.m. tomorrow (Feb. 25) via Zoom.
Lan (Emily) Zhang (ECE) will present “Augmenting Radio Environments for better Wireless Ecosystems.”
Zhang is a member of the Institute of Computing and Cybersystems’s (ICC) Center for Cyber-Physical Systems.
Assistant Professor Sidike Paheding, Applied Computing, has been awarded a one-year MSGC Research Seed Grant for his project, “Monitoring Martian landslides using deep learning and data fusion.”
Professor Thomas Oommen, Geological and Mining Engineering and Sciences, is Co-PI of the project. The grant will support part-time employment of two students during the award period.
This grant is supported in part by funding provided by the National Aeronautics and Space Administration (NASA), under award number 80NSSC20M0124, Michigan Space Grant Consortium (MSGC).
The MSGC Research Seed Grant Program supports junior faculty and research scientists at MSGC affiliate institutions. The program also helps mid-career and senior faculty develop new research programs. The objective of this program is to allow award recipients to develop the research expertise necessary to propose research activities in new areas to other federal or nonfederal sources.
The 2020-2021 CTL Instructional Award for Innovative or Out of Class Teaching is being presented to two instructors, and Assistant Professor Leo Ureel, Computer Science, and Libby Meyer, senior lecturer, Visual and Performing Arts.
Ureel was nominated in recognition of his “student-centric efforts which have increased retention and diversified the cohort of first-year computing students.”
Ureel’s presentation, “Three course innovations to support communication,” will be presented at 3:30 p.m. on Thursday, February 18, 2021, as part of the CTL Instructional Award Presentation Series.
Link here to register for the event.
Ureel is a member of the Institute of Computing and Cybersystems’s (ICC) Computing Education Center.
Meyer’s presentation, “Beyond Carrots and Sticks: Mastery Based Grading and Narrative Assessment” will also be presented on February 18.
During spring 2017, academic deans were asked to begin recognizing instructors making contributions in these areas as part of the Deans’ Teaching Showcase, effectively nominating them for instructional awards.
CTL and Provost’s office members along with previous awardees then select one individual in each category from a pool composed of the Showcase and those nominated to the Academy of Teaching Excellence.
Three course innovations to support communication Introductory courses present many communication challenges between faculty and first year students. In this context, we discuss three innovations used in our introductory computer science courses.
The first is the use of Snap, a high-level, visual programming language, as a form of pseudocode during the first five weeks of the course to build student vocabulary and problem solving skills before tackling programming in Java.
The second is a Code Critiquer developed as a Canvas plugin to provide immediate guidance and feedback to students when they submit their programming assignments.
The third is a grade visualization tool that helps students understand their current performance in the course and project a range that will contain their final grade. While not everyone teaches introductory computer science, we discuss how these or similar innovations and tools might apply to your course.
Dr. Vijay Garg, University of Texas Austin, will present a lecture on February 19, 2021, at 3:00 p.m. The lecture is hosted by the Department of Computer Science.
Vijay Garg is a Cullen Trust Endowed Professor in the Department of Electrical & Computer Engineering at The University of Texas at Austin. He received his Ph.D. in computer science at the University of California at Berkeley and B. Tech. in computer science at IIT, Kanpur.
His research interests are in distributed computing, discrete event systems and lattice theory. He is the author of “Elements of Distributed Computing” (Wiley, 2002), “Introduction to Lattice Theory with Computer Science Applications” (Wiley, 2015), and “Modeling and Control of Logical Discrete Event Systems” (Springer, 2012). He is an IEEE Fellow.
Applying Predicate Detection to Discrete Optimization Problems
We present a method to design parallel algorithms for the constrained combinatorial optimization problems. Our method solves and generalizes many classical combinatorial optimization problems including the stable marriage problem, the shortest path problem and the market clearing price problem.
These three problems are solved in the literature using Gale-Shapley algorithm, Dijkstra’s algorithm, and Demange, Gale, Sotomayor algorithm. Our method solves all these problems by casting them as searching for an element that satisfies an appropriate predicate in a distributive lattice. Moreover, it solves generalizations of all these problems — namely finding the optimal solution satisfying additional constraints called lattice-linear predicates.
For stable marriage problems, an example of such a constraint is that Peter’s regret is less than that of Paul. Our algorithm, called Lattice-Linear Predicate Detection (LLP) can be implemented in parallel with without any locks or compare-and-set instructions. It just assumes atomicity of reads and writes.
In addition to finding the optimal solution, our method is useful in enumerating all constrained stable matchings, and all constrained market clearing price vectors. The talk is an extended version of a paper that appeared in ACM SPAA’20.
Assistant Professor Weihua Zhou, Applied Computing, will present a Chemistry department seminar on Friday, February 12, 2021, from 3:00 to 4:00 p.m. He will present ” Artificial intelligence for medical image analysis: our approaches.”
A publication by Associate Professor Yakov Nekrich, Computer Science, has been accepted to the 53rd Annual ACM Symposium on Theory of Computing (STOC).
The paper, “Optimal-Time Dynamic Planar Point Location in Connected Subdivisions,” describes an optimal-time solution for the dynamic point location problem and answers an open problem in computational geometry.
The data structure described in the paper supports queries and updates in logarithmic time. This result is optimal in some models of computation. Nekrich is the sole author of the publication.
The annual ACM Symposium on Theory of Computing (STOC), is the flagship
conference of SIGACT, the Special Interest Group on Algorithms and
Computation Theory, a special interest group of the Association for
Computing Machinery (ACM).
by Graduate Student Government
Registration for this year’s virtual Graduate Research Colloquium (GRC) is open. Due to the continuation of the SARS-CoV-19 pandemic, the GRC will be held virtually on Thursday and Friday, April 1and 2.
The GRC is a great opportunity to work on your presentation skills and prepare for upcoming conferences. Students are free to give an oral presentation, a poster talk, or both. All talks will be scored by judges from the same field as the presenter.
Cash prizes are available for the top three places in both oral and poster presentations (1st – $300, 2nd – $200, and 3rd – $100). Registration closes Tuesday March 2, at 11:59 PM. Register today.
Poster presentations will take place in a pre-recorded video style. The deadline for video submission is Monday, March 22. A short Q&A session will take place with judges between 4-6 p.m. on April 1. Oral presentations are limited to 12 minutes plus a Q&A session.
The GRC will be capped off with a virtual awards ceremony. All participants and judges are invited to attend. The ceremony will be held on April 2, from 5-7 pm. Full information can be found on our website.
Feel free to contact Sarvada Chipkar if you have any questions or concerns.
The Institute of Computing and Cybersystems will present a Distinguished Lecture by Dr. Alina Zare on Friday, April 16, 2021, at 3:00 p.m.
Her talk is titled, “Multiple Instance Learning for Plant Root Phenotyping.”
Dr. Zare is a professor in the Electrical and Computer Engineering department at University of Florida. She teaches and conducts research in the areas of pattern recognition and machine learning.
Multiple Instance Learning for Plant Root Phenotyping
In order to understand how to increase crop yields, breed drought tolerant plants, investigate relationships between root architecture and soil organic matter, and explore how roots can play in a role in greenhouse gas mitigation, we need to be able to study plant root systems effectively. However, we are lacking high-throughput, high-quality sensors, instruments and techniques for plant root analysis. Techniques available for analyzing root systems in field conditions are generally very labor intensive, allow for the collection of only a limited amount of data and are often destructive to the plant. Once root data and imagery have been collected using current root imaging technology, analysis is often further hampered by the challenges associated with generating accurate training data.
Most supervised machine learning algorithms assume that each training data point is paired with an accurate training label. Obtaining accurate training label information is often time consuming and expensive, making it infeasible for large plant root image data sets. Furthermore, human annotators may be inconsistent when labeling a data set, providing inherently imprecise label information. Given this, often one has access only to inaccurately labeled training data. To overcome the lack of accurately labeled training, an approach that can learn from uncertain training labels, such as Multiple Instance Learning (MIL) methods, is required. In this talk, I will discuss our team’s approaches to characterizing and understanding plant roots using methods that focus on alleviating the labor intensive, expensive and time consuming aspects of algorithm training and testing.
Dr. Zare earned her Ph.D. in December 2008 from the University of Florida. Prior to joining the faculty at the University of Florida in 2016, she was a faculty member at the University of Missouri.
Zare’s research has focused primarily on developing machine learning and pattern recognition algorithms to autonomously understand and process non-visual imagery. Her research work has included automated plant root phenotyping using visual and X-ray imagery, 3D reconstruction and analysis of X-ray micro-CT imagery, sub-pixel hyperspectral image analysis, target detection and underwater scene understanding using synthetic aperture sonar, LIDAR data analysis, Ground Penetrating Radar analysis, and buried landmine and explosive hazard detection.
The Department of Computer Science will present a lecture by Dr. Elizabeth Veinott on Friday, February 12, 2021, at 3:00 p.m.
Veinott is an associate professor in the Cognitive and Learning Sciences department. She will present, “Beyond the system interface: Using human-centered design to support better collaborative forecasting.”
Elizabeth Veinott is a cognitive psychologist working in technology-mediated environments to improve decision making, problem solving and collaboration. She directs Michigan Tech’s Games, Learning and Decision Lab and is the lead for the Human-Centered Computing group of Michigan Tech’s Institute of Computing and Cybersystems (ICC).
She has been active in the ACM’s SIGCHI and on the conference organizing committees for CHI Play and CSCW. Prior to joining Michigan Tech in 2016, she worked as a principal scientist in an industry research and development lab and as a contractor at NASA Ames Research Center. Her research has been funded by NIH, Army Research Institute, Army Research Lab, Air Force Research Laboratory, and IARPA.
Teams use technology to help them make judgments in a variety of operational environments. Collaborative forecasting is one type of judgment performed by analyst teams in weather, business, epidemiology, and intelligence analysis. Research related to collaborative forecasting has produced mixed results.
In her talk, Veinott will describe a case of using cognitive task analysis to develop and evaluate a new forecast process and tool. The method captured analysts’ mental models of game-based forecasting problems, and allowed the process to co-evolve with the system design. The tool was tested in a simulation environment with expert teams conducting analyses over the course of hours and compared to a control group. Challenges and lessons learned will be discussed, including implications for human-centered design of collaborative tools.
A scholarly paper co-authored by Assistant Professor Sidike Paheding, Applied Computing, is one of two papers to receive the 2020 Best Paper Award from the open-access journal Electronics, published by MDPI.
The paper presents a brief survey on the advances that have occurred in the area of Deep Learning.
Paheding is a member of the Institute of Computing and Cybersystems’ (ICC) Center for Data Sciences (DataS).
Co-authors of the article, “A State-of-the-Art Survey on Deep Learning Theory and Architectures,” are Md Zahangir Alom, Tarek M. Taha, Chris Yakopcic, Stefan Westberg, Mst Shamima Nasrin, Mahmudul Hasan, Brian C. Van Essen, Abdul A. S. Awwal, and Vijayan K. Asari. The paper was published March 5, 2019, appearing in volume 8, issue 3, page 292, of the journal.
Papers were evaluated for originality and significance, citations, and downloads. The authors receive a monetary award , a certificate, and an opportunity to publish one paper free of charge before December 31, 2021, after the normal peer review procedure.
Electronics is an international peer-reviewed open access journal on the science of electronics and its applications. It is published online semimonthly by MDPI.
MDPI, a scholarly open access publishing venue founded in 1996, publishes 310 diverse, peer-reviewed, open access journals.
In recent years, deep learning has garnered tremendous success in a variety of application domains. This new field of machine learning has been growing rapidly and has been applied to most traditional application domains, as well as some new areas that present more opportunities. Different methods have been proposed based on different categories of learning, including supervised, semi-supervised, and un-supervised learning. Experimental results show state-of-the-art performance using deep learning when compared to traditional machine learning approaches in the fields of image processing, computer vision, speech recognition, machine translation, art, medical imaging, medical information processing, robotics and control, bioinformatics, natural language processing, cybersecurity, and many others.
This survey presents a brief survey on the advances that have occurred in the area of Deep Learning (DL), starting with the Deep Neural Network (DNN). The survey goes on to cover Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), including Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU), Auto-Encoder (AE), Deep Belief Network (DBN), Generative Adversarial Network (GAN), and Deep Reinforcement Learning (DRL). Additionally, we have discussed recent developments, such as advanced variant DL techniques based on these DL approaches. This work considers most of the papers published after 2012 from when the history of deep learning began.
Furthermore, DL approaches that have been explored and evaluated in different application domains are also included in this survey. We also included recently developed frameworks, SDKs, and benchmark datasets that are used for implementing and evaluating deep learning approaches. There are some surveys that have been published on DL using neural networks and a survey on Reinforcement Learning (RL). However, those papers have not discussed individual advanced techniques for training large-scale deep learning models and the recently developed method of generative models.
