 |
 |
 |
The Department of Energy’s Artificial Intelligence and Technology Office
along with The National Nuclear Security Agency’s Office of Proliferation (NNSA) presents its 2023 Challenge:
Finding Indicators of Nuclear Activity in Open-Source Images using AI/ML
Today, limited quantities of real images relevant to some types of nuclear activities exist in open sources – too few to train robust computer vision models for the wide variance required in the operational environment. In anticipation of extremely limited data environments, computer vision models must be able to train exclusively on synthetic, computer-generated images, while achieving high performance on real-world images.
SME
Siddha Ganju
Siddha Ganju is an AI researcher who Forbes featured in their 30 under 30 list. She is an AI researcher at NVIDIA who leads the development of neural networks optimizations for medical instruments and life sciences companies. She leads, designs, and helps deploy real-time and low power AI applications to enable impact of AI in traditional sciences and domains. She held the title of Self-Driving Architect at Nvidia, working towards stable and scalable training of neural networks on very large data centers, and utilizing simulation to validate the neural networks. As an AI Advisor to NASA FDL, she helped build an automated meteor detection pipeline for the CAMS project at NASA, which resulted in the discovery of multiple meteors. Previously at Deep Vision, she developed deep learning models for resource constraint edge devices. Her work ranges from Visual Question Answering to Generative Adversarial Networks to gathering insights from CERN’s petabyte-scale data and has been published at top-tier conferences including CVPR and NeurIPS. She has served as a featured jury member in several international tech competitions including CES. She is also the author of O’Reilly’s Practical Deep Learning for Cloud, Mobile and Edge. As an advocate for diversity and inclusion in technology, she speaks at schools and colleges to motivate and grow a new generation of technologies from all backgrounds. Siddha is a growing horticulturist and enjoys hiking.
TEAM ADVISORS
Robert Alvarez
Advisor
Robert Alvarez is a seasoned data scientist known for his problem-solving acumen. At Intel, he used his knowledge to tackle problems in data center optimization using cluster analysis, enriched market sizing models by implementing sentiment analysis from social media feeds, and improved data-driven decision making in one of their top 5 global supply chains. At Tamr, he built models to unify large amounts of messy data across multiple silos for some of the largest corporations in the world. At Frontier Development Lab he has worked on challenges for both NASA and the DOE (super resolution of the sun, flood prediction, and computer vision models to detect radiation in urban environments). In his role at Podium Education, Robert has equipped students with data analysis skills to enhance their job readiness post-graduation. He earned a PhD in Applied Mathematics from Arizona State University where his research spanned image reconstruction, dynamical systems, mathematical epidemiology, and oncology. In his spare time, he is a rum aficionado, avid traveler, and eater of all things coconut.
Bennie Lewis
Advisor
Bennie Lewis is a LM Associate Fellow in the Advanced Technology Center at Lockheed Martin Space Sunnyvale, Palo Alto California with a focus on AI, Deep Learning/Machine Learning, AR/VR, Embedded Systems, Cognitive Systems, Space Systems, Human Machine Interaction, Robotics, Astrodynamics, and Space Situational/Domain Awareness. Prior to his current role, he was a Software Engineer for Lockheed Martin Missiles and Fire Control Orlando Florida focusing on Fixed Wing and Missile defense systems. He has a B.S. and M.S in Computer Engineering as well as a Ph.D. in Computer Engineering with a focus in AI/ML, robotics, and intelligent systems from the Department of Electrical Engineer and Computer Science at the University of Central Florida working in the Intelligent Agents Laboratory. A second PhD in Aerospace Engineering with a focus on AI/ML for Astrodynamics, Controls, and Space Robotics in the Department of Mechanical and Aerospace Engineering at the University of Central Florida working in the Astrodynamics and Space Robotics Laboratory.
Zoe Gastelum
DOE Advisor
Zoe Gastelum is a Principal Member of the Technical Staff in the International Safeguards and Engagements Department at Sandia National Laboratories. Her research foci include data analysis techniques, methods, and implications for international nuclear safeguards verification, the performance of human-information safeguards systems, and challenges and opportunities of emerging technologies for international safeguards. Prior to joining Sandia, she spent five years as a nonproliferation scientist at Pacific Northwest National Laboratory and two years as an open-source information analyst in the International Atomic Energy Agency’s Department of Safeguards.
Zoe has a B.A. in Political Science from the University of New Mexico, where she graduated summa cum laude with honors in International Studies. She holds a M.A. in International Security from the University of Texas at Austin’s Teresa Lozano Long Institute of Latin American Studies and has a graduate certificate in International Nuclear Law from the Université Montpellier 1 in France. She is currently earning a Ph.D. in Public Policy with a focus in Science & Technology Policy at the Georgia Institute of Technology, where her research focuses on artificial intelligence policy for international safeguards.
Timothy Shead
DOE Advisor
Lakshman Prasad
DOE Advisor
Dana Grisham
DOE Advisor
RESEARCHERS
Jaden Hastings
Lead Researcher
Jaden J. A. Hastings is an expeditionary scientist specializing in the study of biotic life within the context of extreme and isolated conditions. Her passion for exploration drives her to descend deep into lava tubes, traverse thousands of miles of desert, or sail across The Drake Passage to Antarctica. After completing her undergraduate studies at New York University, she went on to complete graduate degrees from Harvard and Oxford in Biology and Bioinformatics, respectively, followed by doctoral research at the University of Melbourne focused on the development of machine-driven mechanisms for biological evolution.
Jaden is also an internationally exhibited artist and earned an MFA from Central Saint Martins in London. Last year, Jaden served as a Researcher for the Astrobiology Team for FDL 2022, which focused on developing an ML pipeline for agnostic biosignature detection that could be deployed on an interplanetary probe, such as Dragonfly.
David Nyarko
Researcher
David Nyarko is a highly accomplished engineer specializing in robotics and artificial intelligence. Currently pursuing a Doctor of Electrical Engineering at Morgan State University, he holds a BSc. in Engineering Physics from the University of Cape Coast, Ghana. With expertise in machine learning, data modeling, signal processing, and programming, David has worked on various projects involving robot development, cross-platform software development, microcontroller programming, and game development. As a Graduate Research Fellow, he is actively involved in cutting-edge research at the Data Engineering and Predictive Analytics Lab at Morgan State University, where he builds object detection and classification models and implements text recognition on video streams while contributing to an automatic wheelchair project in the lab.
As the CEO/Co-founder of SwiftDynamics Africa, David has led groundbreaking initiatives, including the development of a Robot Operating System and Ghana’s first AI social robot. Each step of his journey is geared towards empowering the next generation of tech giants from Africa. When David is not doing research, he loves to skate or play basketball on rollerblades.
Argho Sarkar
Researcher
Argho Sarkar received a B.S. in Applied Statistics from the University of Dhaka, Bangladesh, in 2018. He received his M.S. in Information Systems from the University of Maryland, Baltimore, in 2022 where he is currently pursuing a Ph.D. in Information Systems. His research mainly focuses on developing AI and ML techniques for social good. He is doing his Ph.D. research on developing vision-language-based multimodal deep learning algorithms such as visual question answering and image captioning for remote sensing contexts with the aim of providing rapid response and recovery solutions following a disaster. He has multiple publications at top-tier conferences and in journals, including ICML, ICIP, TGRS, and IGARSS. He was one of the organizers of the FloodNet Challenge at CVPR 2021. He has experience working as a statistician in the public health sector in Bangladesh and as an applied science intern at Amazon Web Services (AWS) in the U.S. Besides research, Argho loves to watch movies, travel, and cook with family and friends.
Pugazhenthi Sivasankar
Researcher
Pugazhenthi Sivasankar is a third year Ph.D. student at the University of Central Florida, USA. His research interests include astrodynamics and functional approximation. Being an autistic student from the south of India, he has always been interested in math and physics. He went through a major cultural and academic shift when he moved to the Netherlands for his Masters in Spaceflight, which led him to further academic opportunities in Belgium, China, Sweden, and the US.
Pugal values individual technical brilliance as well as being a team player. During his Bachelors in India, he was involved in technical activities of his department and social service activities of his college. During his graduate school in the Netherlands, he worked as a Teaching Assistant to two groups of multi-cultural, multi-linguistic students who did the preliminary design for a Mars orbiter. The following year he worked as a thesis intern at Qinetiq Space, Belgium and attended the 2nd ISSI-BJ Space Science School in Sanya, China. He continues to be a team player in his Ph.D. research which is funded by Lockheed Martin. His paper on validation of uncertainty quantification of space debris presented at the 3rd ICSSA won the Best Student Award.
GUEST RESEARCHER
Andria Gazda
Andria Gazda is a student researcher with a multidisciplinary background in biology, computer science, and mathematics with an expertise in statistics. She has two A.S. degrees in Mathematics and in Computer Science and holds a B.S. in Biological Sciences from the University of California, Irvine. She excels at transdisciplinary teamwork. Andria was one of the competitively selected students who represented the U.S. for the 2022 Japan-America Student Conference (JASC) established in 1934, where she spent her summer studying and analyzing Japan-U.S. relations while visiting four diverse regions in the host country, alongside her Japanese contemporaries.
Her research experience includes an internship in the Cystic Fibrosis Laboratory at Johns Hopkins in Baltimore, MD. As a Research Fellow in UCI’s ID-SURE program, she was the two-year Team Lead and Coordinator for the Rocky Flats data science research project. The results helped to secure health care coverage for former workers of the Rocky Flats plant by uncovering documentation of radiation exposure. She is the co-author of the book Applied Calculus for Business, Economics, and Social Sciences –(Arismendi-Pardi, Gazda, Fritsche). Andria enjoys baking, online gaming with her family, and poring over scholarly databases to research questions for fun.