Multidisciplinary Research and Education on Big Data + High-Performance Computing + Atmospheric Sciences

Part of NSF Initiative on Workforce Development for Cyberinfrastructure (CyberTraining)

CyberTraining Spring 2019


Introduction

We conducted online training to 17 participants (5 from computing discipline, 6 from physics discipline, 6 from mathematics discipline). Among them, we have 1 master student, 13 PhD students, 3 junior faculty/scientists. Besides the 4 instructors/mentors and 3 dedicated TAs, we also invited 2 external mentors. Each team successfully produced a technical report. Reports are being extended for peer-reviewed conference/journal publications. Evaluation shows the training program is very useful to participants, especially the program provides a unique experience for them to work with peers in other disciplines on a research project.


Projects

Team 1 Project

Team 2 Project

Team 3 Project

Team 4 Project

Team 5 Project


Team 1 Project: Assessing water budget sensitivity to precipitation forcing errors in Potomac river basin using the VIC hydrologic model

Team members:
Reetam Majumder, Department of Mathematics and Statistics, UMBC
Redwan Walid, Department of Information Systems, UMBC
Jianyu Zheng, Department of Physics, UMBC
Graduate Assistant:
Carlos Barajas, Department of Mathematics and Statistics, UMBC
Pei Guo, Department of Information Systems, UMBC
Chamara Rajapakshe, Department of Physics, UMBC
Faculty Mentor:
Aryya Gangopadhyay, Department of Information Systems, UMBC
Matthias K. Gobbert, Department of Mathematics and Statistics, UMBC
Jianwu Wang, Department of Information Systems, UMBC
Zhibo Zhang, Department of Physics, UMBC
Deliverables:
Implementation Source Code at Github Repository
Presentation Slides
Technical Report
Peer reviewed publication: In preparation

back to top

Team 2 Project: Evaluation of Data-Driven Causality Discovery Approaches among Dominant Climate Modes

Team members:
Steve Hussung, Department of Mathematics, Indiana University, Bloomington
Suhail Mahmoud, Computational Science Department, University of Texas at El Paso
Akila Sampath, Department of Atmospheric Sciences, University of Alaska, Fairbanks
Mengxi Wu, Department of Earth, Environmental and Planetary Sciences, Brown University
Graduate Assistant:
Pei Guo, Department of Information Systems, UMBC
Faculty Mentor:
Jianwu Wang, Department of Information Systems, UMBC
Deliverables:
Implementation Source Code at Github Repository
Presentation Slides
Technical Report
Peer reviewed publication: In preparation

back to top

Team 3 Project: An approach to tuning hyperparameters in parallel: A performance study using climate data

Team members:
Charlie Becker, Department of Geosciences, Boise State University
Will D. Mayfield, Department of Mathematics, Oregon State University
Sarah Y. Murphy, Department of Civil and Environmental Engineering, Washington State University
Bin Wang, Department of Mathematics, Hood College
Graduate Assistant:
Carlos Barajas, Department of Mathematics and Statistics, UMBC
Faculty Mentor:
Matthias K. Gobbert, Department of Mathematics and Statistics, UMBC
Deliverables:
Implementation Source Code at Github Repository
Presentation Slides
Technical Report
Peer reviewed publication: In preparation

back to top

Team 4 Project: Deep Learning Based Mineral Dust Detection and Feature Selection

Team members:
Ping Hou, School for Environment and Sustainability, University of Michigan
Peng Wu, Department of Hydrology and Atmospheric Sciences, University of Arizona
Graduate Assistant:
Pei Guo, Department of Information Systems, UMBC
Faculty Mentor:
Aryya Gangopadhyay, Department of Information Systems, UMBC
Deliverables:
Implementation Source Code at Github Repository
Presentation Slides
Technical Report
Peer reviewed publication: In preparation

back to top

Team 5 Project: Dust Detection in Satellite Data using Convolutional Neural Networks

Team members:
Changjie Cai, Department of Occupational and Environmental Health, University of Oklahoma
Jangho Lee, Department of Atmospheric Sciences, Texas A&M University
Yingxi Rona Shi, USRA and NASA GSFC Climate and Radiation Laboratory
Camille Zerfas, School of Mathematics and Statistics, Clemson University
Graduate Assistant:
Pei Guo, Department of Information Systems, UMBC
Faculty Mentor:
Zhibo Zhang, Department of Physics, UMBC
Deliverables:
Implementation Source Code at Github Repository
Presentation Slides
Technical Report
Peer reviewed publication: Jangho Lee, Yingxi R. Shi, Changjie Cai, Pubu Ciren, Jianwu Wang, Aryya Gangopadhyay, Zhibo Zhang. 2021. "Machine Learning Based Algorithms for Global Dust Aerosol Detection from Satellite Images: Inter-Comparisons and Evaluation" Remote Sens. 13, no. 3: 456. https://doi.org/10.3390/rs13030456.

back to top

Some testimonials from anonymous evaluations

  • “It is one of the best educational and research experience I have in my educational life.”
  • “Although it was an online course, it taught me many topics which I haven't learned in a real-life class. It is highly recommended to anyone who wants to thrive in his/her educationcal and research career.”
  • “Great way to have a jump start on learning high efficiency computing and machine learning in an interdisciplinary way, this is the course for you.”
  • “I recommend the CyberTraining course because you get to learn a lot of relevant and useful topics that help to expand your research. The projects at the end are very interesting, and you have the opportunity to publish if you want.”
  • “You get trained on comprehensive skill set. The courses are well organized and the mentors/TAs are willing to help.”
  • “Excellent framework to contribute from your respective field, learn from the experts of others, and unite to solve challenging problems. I've become a better programmer and better data scientist as a result.”