Aims and Scope
The significant advancements in software and hardware technologies stimulated the prosperities of the domains in spatial computing and deep learning algorithms, respectively. Recent breakthroughs in the deep learning field have exhibited outstanding performance in handling data in space and time in specific domains such as image, audio, and video. Meanwhile, the development of sensing and data collection techniques in relevant domains have enabled and accumulated large scale of spatiotemporal data over the years, which in turn has led to unprecedented opportunities and prerequisites for the discovery of macro- and micro- spatiotemporal phenomena accurately and precisely. The complementary strengths and challenges between spatiotemporal data computing and deep learning in recent years suggest urgent needs to bring together the experts in these two domains in prestigious venues, which is still missing until now.
This workshop will provide a premium platform for both research and industry to exchange ideas on opportunities, challenges, and cutting-edge techniques of deep learning in spatiotemporal data, applications, and systems.
Topics of Interest: We encourage submissions of papers that fall into (but not limited to) the following three broad categories:
Novel Deep Learning Techniques for Spatial and Spatio-Temporal Data:
Spatial representation learning and deep neural networks for spatio-temporal data and geometric data
Physics-guided and interpretable deep learning for spatial-temporal dataDeep generative models for spatio-temporal data
Deep reinforcement learning for spatio-temporal decision making problems
Novel Applications of Deep Learning Techniques to Spatio-temporal Computing Problems. :
Remote sensing imagery and point cloud analysis in Earth science (e.g., hydrology, agriculture, ecology, natural disasters, etc.)
Deep learning for mobility and traffic data analyticsLocation-based social network data analytics, geosocial media data mining, spatial event prediction and forecasting, geographic knowledge graphs
Learning for biological data with spatial structures (bio-molecule, brain networks, etc.)
Challenges, Opportunities, and Early Progress in Deep Learning for COVID-19
Novel Deep Learning Systems for Spatio-temporal Applications:
Real-time decision-making systems for traffic management, crime prediction, accident risk analysis, etc.
GIS systems using deep learning (e.g., mapping, routing, or Smart city)Mobile computing systems using deep learningGeoAI Cyberinfrastructure for Earth science applicationsInterpretable deep learning systems for spatio-temporal temporal data
Program
(All times are in ET time zone)
Location: Room 204C
| 1:00pm-1:05pm | Opening Remark |
| 1:05pm-1:45pm | Keynote 1 by Dr. Chaowei Phil Yang, GMU |
| 1:45pm-2:35pm | Keynote 2 by Dr. Stefano Ermon, Stanford |
| 2:30pm-3:30pm | Paper Splotlight Presentation: |
| 3:30pm-5:00pm | Panel: The Future of AI for Spatiotemporal Data Science Panelists: Dr. Seung-Jong Park (NSF), Dr. Vandana Janeja (UMBC), Dr. Ray Dos Santos (GRL), and Dr. Jack Cooper (IARPA). |
| 5:00pm-5:05pm | Closing Remark |
Keynotes Speakers
Abstract:
Climate change and pollutant emissions continue to worsen our breathing air, causing severe health problems for national and global citizens including ~ 10 million premature death annually and many more with relevant disease and longer term health impact. High resolution and fidelity data is desperately needed to support decision making to mitigate such health impact. Several questions that came to decision making need include a) how to better detect and forecast air quality events at community level, b) how to identify the correlation spatiotemporally between various breathing disease and the air quality patterns, c) how to develop relevant policies and a rapid response system that can help mitigate the impact of air pollution to the vulnerable communities to improve their resilience and sustainability. Taking these questions, this talk will introduce the popular air quality datasets, processing needs and the opportunities and challenges for machine learning such as research aspects of training datasets, dynamic, causal, multi-model, and multi-source methods.
Speaker Bio:
Chaowei Phil Yang is Professor of GIScience at George Mason University. His research focuses on utilizing spatiotemporal principles/patterns to optimize computing infrastructure to support science discoveries and engineering development. He has been funded as PI by multiple resources such as NSF and NASA with over $20M expenditures. He published over 300 papers, edited six books and 10+ special issues for international journals. His publications have been among the top five cited and read papers of IJDE and CEUS. His PNAS spatial computing definition paper was captured by Nobel Intent Blog in 2011. He has led several teams who impact GIScience profoundly. For example, he led a 3-member team who developed GeoServNet at Univ. of Calgary. The software was evolved into the core of GeoTango, which was purchased by Microsoft as the origin of Virtual Earth/Bing Maps. He furthers this innovation process by encouraging and creating startups with his students and collaborators. He has placed eight tenure line Professors in the U.S. and 15 Associate or Full Professors internationally including top GIScience departments such as Penn State, Univ. of Wisconsin- Madison, Arizona State, and Wuhan Univ. He also served in many public positions, such as the President of the CPGIS (2004-2005) and the chair of UCGIS research committee (2012-2014). He received multiple national and international awards, such as the Environment Stewardship Award in 2009 from President Obama. He founded NSF I/UCRC for spatiotemporal thinking, computing and applications with a group of international leaders from UCSB, Harvard, and GMU. The spatiotemporal innovation center receives over $2M/year research funding in collaboration with agencies and industry to build the national and international spatiotemporal infrastructure.
Abstract:
Data streams satellites, phones, and other digital devices contain a wealth of information relevant to sustainability goals. Modern AI techniques have the potential to yield accurate, inexpensive, and highly scalable models to inform research, policy, and business decisions. A key challenge, however, is the lack of large quantities of labeled data that often characterize successful machine learning applications. In this talk, I will present approaches for learning useful spatio-temporal models in contexts where labeled training data is scarce or not available at all. I will show applications to predict key socio-eocnomic indicators, monitor agricultural productivity, map infrastructure access, predict population health measures in data poor areas, and discuss how these insights are being used to improve decisions on the ground.
Speaker Bio:
Stefano Ermon is an Associate Professor of Computer Science in the CS Department at Stanford University, where he is affiliated with the Artificial Intelligence Laboratory, and a fellow of the Woods Institute for the Environment. His research is centered on techniques for probabilistic modeling of data and is motivated by applications in the emerging field of computational sustainability. He has won several awards, including Best Paper Awards (ICLR, AAAI, UAI and CP), a NSF Career Award, ONR and AFOSR Young Investigator Awards, a Sony Faculty Innovation Award, a Hellman Faculty Fellowship, Microsoft Research Fellowship, Sloan Fellowship, and the IJCAI Computers and Thought Award. Stefano earned his Ph.D. in Computer Science at Cornell University in 2015.
Panel: The Future of AI for Spatiotemporal Data Science
Jack Cooper Intelligence Advanced Research Projects Activity (IARPA) |
1. Dr. Seung-Jong Jay Park is the Dr. Fred H. Fenn Memorial Professor of Computer Science and Engineering at Louisiana State University where he has worked in cyberinfrastructure development for large-scale scientific and engineering applications since 2004. He received Ph.D. in the school of Electrical and Computer Engineering from the Georgia Institute of Technology (2004). He has performed interdisciplinary research projects including (1) big data & deep learning research including developing software frameworks for large-scale science applications and (2) cyberinfrastructure development using cloud computing, high-performance computing, and high-speed networks. Those projects have been supported by federal and state funding programs from NSF, NASA, NIH, ONR, AFRL, etc. He received IBM faculty research awards between 2015-2017. He also served an associate director for the Center for Computation and Technology of LSU between 2016-2018. Since 2021 he has served at the U.S. National Science Foundation (on leave from LSU) as a program director managing research support programs, such as Cyberinfrastructure for Sustained Scientific Innovation (CSSI), Principles and Practice of Scalable Systems (PPoSS), Computational and Data-Enabled Science and Engineering (CDS&E), OAC Core, etc.
2. Dr. Vandana Janeja is Professor and Chair of the Information Systems department at the University of Maryland Baltimore County (UMBC) and director of iHARP, an NSF HDR Institute for Harnessing Data and Model Revolution in the Polar Regions. Her research is in the area of data science focusing on spatio-temporal mining, data heterogeneity across multi-domain datasets. She heads the Multi Data lab at UMBC which brings together important societal projects such as climate change, ethics in data science, misinformation detection and security through the lens of her research in data science. She is member of the UMBC ADVANCE Executive committee focusing on diversity in STEM and a member of the ADVANCE leadership cohort (2020-2021), an ELATES at Drexel leadership fellow (2021-2022) and a UMBC innovation fellow (2020-2022) advancing the ideas of including ethics in data science. She served as an expert at NSF supporting data science activities in the CISE directorate (2018-2021) and AAAS S&TP fellow (2017-2018), where she helped with the visioning and coordination of cross directorate activities for Harnessing the Data Revolution Big Idea at NSF including Data Science Corps and Open Knowledge Network among others and, related activities in the CISE directorate including Cloud Access. Her research has been funded through federal, state and private organizations including NSF, U.S. Army Corps of Engineers, MD State Highway Administration, CISCO. She holds a Ph.D. in Information Technology from Rutgers University. She completed her MBA from Rutgers University and MS in Computer Science from New Jersey Institute of Technology.
3. Dr. Ray Dos Santos is a computer scientist and program manager at the Geospatial Research Lab of the U.S. Army Corps of Engineers. For the past 11 years, Ray has worked on various projects related to threat detection, disaster relief, text analysis, and entity networks. His research interests include semantic extraction from text and images, graph networks, spatial data mining, and visualization. He maintains close cooperation with academia and Government offices such as the US Geological Survey, and the Department of Defense. Ray received his PhD and Master's in Computer Science from Virginia Tech, and completed his undergraduate degree from the University of South Florida.
4. Dr. Jack Cooper joined IARPA in October of 2020. In his current assignment as a Program Manager, Dr. Cooper focuses on areas of scientific research that include remote sensing, computer vision, trajectory analytics, and large-scale microsimulation. He is IARPA's Program Manager for three programs focused on automating and scaling the IC's ability to detect and characterize human activity in a variety of mediums. These programs are SMART, which seeks to automate the broad area search for a specific activity in satellite imagery; DIVA which aims to automatically find and prioritize activities of interest in massive volumes of security video; and HAYSTAC which will automatically identify anomalous activity by deeply modeling normal human movement in large-scale trajectory data. Prior to joining IARPA, Dr. Cooper worked at the National Geospatial-Intelligence Agency (NGA) in the Research Directorate, where he was a Senior Staff Scientist for Predictive Analytics. There Dr. Cooper created and executed significant programs to automate Motion GEOINT processing and analysis, resulting in the transfer of multiple analytics into operations in both NGA and IC Partner enterprises. During his time at NGA, Dr. Cooper also built research programs to prepare the agency for future satellite systems and data streams through field experiments, simulation, and creation of new processing algorithms and architectures. Dr. Cooper graduated from Clemson University in 2012 with a Ph. D and Master's degree in Mathematical Sciences. He previously graduated from the University of Maryland in 2007 with a Bachelor’s degree in Mathematics and Government and Politics.
Accepted Papers
1. EAST: An Enhanced Automated Machine Learning Library for Spatio-Temporal Forecasting
Songyu Ke, Li Song, Kainan Bao, Zheyi Pan, Junbo Zhang, and Yu Zheng.
2. Periodic Residual Learning for Crowd Flow Forecasting
Chengxin Wang, Yuxuan Liang, and Gary Tan.
Xiaolin Chang, Shaofu Lin, and Xiliang Liu.
Gary Doran, Serina Diniega, Steven Lu, Mark Wronkiewicz, Kiri Wagstaff, and Jacob Widmer.
Workshop Co-Chairs
Publicity Chair
Siyi Gu, Emory University
Jason Jin, Emory University
Program Committee
Arnold Boedihardjo, DigitalGlobe
Wei Wang, Microsoft Research
Ray Dos Santos, Army Corps of Engineers
Chao Zhang, Georgia Tech
Yanjie Fu, UCF
Xuchao Zhang, NEC Lab North America
Yanfang Ye, University of Notre Dame
Yanhua Li, WPI
Jing Dai, Google
Yiqun Xie, University of Maryland
Xiaowei Jia, University of Pittsburgh
Junbo Zhang, JD Digital
Jie Bao, JD Digital
Song Gao, University of Wisconsin, Madison
Jingyuan Wang, Beihang University
Lexie Yang, ORNL
Borko Furht, Florida Atlantic University
Taghi Khoshgoftaar, Florida Atlantic University
Senzhang Wang, Central South University
Paper Submission
Important Dates: (all due Midnight Anywhere on Earth).
Paper Submission: June 9, 2022
Paper Review Begins: June 10, 2022
Notification of Acceptance: June 26, 2022
Camera-ready Papers: July 4, 2022
Workshop Date: August 15th, 2022
The workshop welcomes the two types of submissions
Full research papers – up to 9 pages (8 pages at most for the main body and the last page can only hold references)
Vision papers and short system papers - up to 5 pages (4 pages at most for the main body and the last page can only hold references)
All manuscripts should be submitted in a single PDF file including all content, figures, tables, and references, following the format of KDD conference papers. Paper submissions need to include author information (review not double blinded).
Papers should be submitted at: https://easychair.org/conferences/?conf=deepspatial22
Concurrent submissions to other journals and conferences are acceptable. Accepted papers will be presented as posters or short talks during the workshop and published on the workshop website. Besides, a small number of accepted papers may be selected to be presented as contributed talks. As a tradition, accepted workshop papers are NOT included in the ACM Digital Library. The authors maintain the copyright of their papers.
Contact Information
Zhe Jiang, zhe.jiang@ufl.edu, Tel: (352) 294-6659
Liang Zhao, liang.zhao@emory.edu , Tel: (703) 993 5910
Xun Zhou, xun-zhou@uiowa.edu, Tel: (319) 384-3335
Junbo Zhang, msjunbozhang@outlook.com
Robert Stewart, stewartrn@ornl.gov, Tel: (865) 574-7646
Copyright © All rights reserved | This template is made with by Colorlib