Embedded Files
JCET Seminar Series Spring 2020
JCET Seminar Series Spring 2020
Application of Artificial Intelligence (AI) in Earth Sciences
Application of Artificial Intelligence (AI) in Earth Sciences
(Feb. 04, 2020 ~ May 15, 2020)
(Feb. 04, 2020 ~ May 15, 2020)
When: Tuesday 11:15 AM ~ 12:15 PM
When: Tuesday 11:15 AM ~ 12:15 PM
Where: ENGR025 (Engineering Building Room 025)
Where: ENGR025 (Engineering Building Room 025)
WebEx info: https://umbc.webex.com/join/jzheng3
WebEx info: https://umbc.webex.com/join/jzheng3
Google Drive Folder:
https://drive.google.com/open?id=1rrasPszAYF5Mnn3tPn4Mt2pL_xBdL7Pw
Artificial intelligence (AI) techniques have brought revolutionary changes to our daily life over the past few years. They are also increasingly used in almost every area of Earth Sciences. The astronomically growing measurements from satellite, ground based and in situ platforms provide the observational basis for training and validating the AI algorithms, which can be fully utilized using the “big-data” techniques.
The objectives of this JCET seminar series are to provide 1) an introductory overview of the basic AI and “big-data” techniques; 2) hand-on experiences with AI based algorithm for satellite remote sensing; 3) an outlook of cutting-edge research of the applications of AI in Earth Sciences.
Lecturers:
Prof. Aryya Gangopadhyay (Department of Information Systems, UMBC)
Prof. Jianwu Wang (Department of Information Systems, UMBC)
Prof. Sanjay Purushothum (Department of Information Systems, UMBC )
Prof. Zhibo Zhang (Department of Physics/JCET, UMBC)
Dr. Chenxi Wang (JCET/UMBC and NASA Goddard Space Flight Center)
Dr. Daniel Miller (JCET/UMBC and NASA Goddard Space Flight Center)
Part I: Basics of AI and big data (6 weeks, Aryya, Jianwu and Sanjay)
02/04/2020 Week #1: Basics of atmospheric remote sensing (Prof. Zhibo Zhang)
Lecture Slide and Recorded Video of Lecture 1:
JCET_seminar_lecture_1.mp4Lecture1_introduction.pptx02/11/2020 Week #2: Foundations of deep learning I (Prof. Aryya Gangopadhyay)
Lecture Slide and Recorded Video of Lecture 2:
JCET_seminar_lecture_2.mp4NeuralNetworks.pptx02/18/2020 Week #3: Foundations of Machine-Learning (Prof. Sanjay Purushotham )
Lecture Slide and Recorded Video of Lecture 3:
JCET_Seminar_lecture_3.mp4JCET_Seminar_Sanjay_ML_talk_1.pdf02/25/2020 Week #4: Basics of Big data (Prof. Jianwu Wang )
Lecture Slide and Recorded Video of Lecture 4:
JCET_seminar_lecture_4.mp4Lecture on Big Data - Jianwu Wang - JCET Seminar Spring 2020.pdfClimate Data Challenges in the 21st Century (2011) - Science.pdfThe Hadoop Distributed File System-2010.pdfChallenges and Opportunities with Big Data.pdfA BIG DATA GUIDE TO UNDERSTANDING CLIMATE CHANGE-2014.pdfMapReduce - A Flexible Data Processing Tool (CACM 2010).pdfDask - Parallel Computation with Blocked algorithms and Task Scheduling-2015.pdf03/03/2020 Week #5: Foundations of Machine-Learning (Prof. Sanjay Purushotham )
The lecture was not recorded due to a technical difficulty. Sorry!
03/11/2020 Week #6: Foundations of deep learning I (Prof. Aryya Gangopadhyay)
CNN.pptxArticle_MachineLearningAndDeepLearning2019.pdfWeek #7: Introduction of basic radiative transfer theory and satellite cloud remote sensing By Dr. Chenxi Wang (JCET/UMBC GSFC/NASA)
JCET_Seminar_lecture_CWang.pptxJCET_seminar_lecture_7.mp4Week #8: Hand-on Experiences: Cloud detection and cloud thermodynamic phase classification using machine-learning based algorithms By Dr. Chenxi Wang (JCET/UMBC GSFC/NASA)
Training Data and Python Code: (see code example at )
JCET_Seminar_lecture2_CWang.pptxJCET_seminar_lecture_8.mp404/07/2020 Week #9 Theoretical basis: Retrieving cloud microphysical properties from polarimetric observations of Research Scanning Polarimeter (RPS) using AI-based algorithms (Dr. Daniel Miller)
JCET_ML_Seminar_DJM.pptxJCET_seminar_lecture_9.mp404/14/2020 Week #10 Hand-on experiences: Retrieving cloud microphysical properties from polarimetric observations of Research Scanning Polarimeter (RPS) using AI-based algorithms (Dr. Daniel Miller NASA GSFC)
JCET_ML_DJM_pt2.pptxJCET_seminar_lecture_10.mp4Exercises and Dataset Access
Exercises and Dataset Access
Tensorflow examples and exercises to try out:
•https://www.tensorflow.org/tutorials
•https://www.tensorflow.org/tutorials/keras/regression
NNv2 Final Data Products
•https://data.giss.nasa.gov/pub/rsp/ORACLES_2016/
Training set data:
•https://oceancolor.gsfc.nasa.gov/fileshare/kirk_knobelspiesse/nn/
•https://oceancolor.gsfc.nasa.gov/fileshare/kirk_knobelspiesse/nn_aca/
4/28/2020 Week #12 Cutting-Edge Research: Combining Experts and Deep Learning to Obtain Insights from NASA Data (Expert Dr. Tianle Yuan NASA GSFC)
Dr. Yuan is currently an Associate Research Scientist in the Climate and Radiation Laboratory of NASA Goddard Space Flight Center, affiliated with UMBC JCET. Dr. Yuan received his Ph.D. from the University of Maryland College Park before he joined the NASA GSFC. His research interests cover a broad array of topics, from Satellite remote sensing; aerosol-cloud-precipitation interactions, to dust variability and climate change, and to Artificial Intelligence and deep learning techniques. For more information about Dr. Tianle Yuan please see his CV and visit his website https://airbornescience.nasa.gov/person/Tianle_Yuan
JCET_seminar_lecture_11.mp4Yuan_expert_machine_UMBC.pptx05/06/2020 Week #13 Cutting-Edge Research: Leveraging Modern AI for the Exploitation of Satellite Data (Expert Dr. Eric Maddy NOAA/NESDIS/STAR)
Dr. Eric S. Maddy received degrees in B.S. physics and mathematics from Frostburg State University, Frostburg, MD in 2001. He received his M.S. and Ph.D. degrees in Atmospheric Physics from the University of Maryland, Baltimore County (UMBC) in 2003 and 2007, respectively. Since 2004 he has been a senior research scientist with Riverside Technology, inc. at NOAA/NESDIS/STAR. During his time at STAR his research focused on the development and analysis of algorithms for deriving temperature, moisture and carbon trace gases from operational IR and MW sounders and imagers in all weather conditions. He has also worked to develop advanced satellite data assimilation and data fusion techniques with a focus on increasing the utilization and effectiveness of satellite data and improving the nation’s premier operational numerical weather prediction (NWP) models. More recently he has applied machine learning and artificial intelligence techniques to remote sounding, satellite calibration, data assimilation and forecasting.
JCET_seminar_lecture_12.mp4200505_UMBC_maddy.pptxReport abuse