HAO Colloquium - Madhulika Guhathakurta, NASA Ames

Applied Artificial Intelligence for Science & Exploration Enabled by Public-Private Partnerships

NASA Frontier Development Laboratory (FDL) is an applied Artificial Intelligence (AI) research partnership focused on challenges in space science and exploration. The recent advances in Artificial Intelligence (AI) capabilities are particularly relevant to NASA science and exploration goals because there is growing evidence that AI techniques can improve our ability to model, understand and predict our environment using the petabytes of data already within NASA archives. The 2019 program, held at the SETI Institute and NASA Ames Research Center focused on AI based science opportunities within the fields of Earth Observation applied to flood prediction and response, Lunar Resources, Astronaut Health, Heliophysics and Space Weather. The program is an 8-week concentrated R&D deep-dive process using the latest developments in machine learning and access to cutting-edge academic research - and enormous compute resources provided by FDL’s commercial AI partners - as well as detailed industry partner case studies, showcasing the potential application of AI to space challenge domains.

NASA FDL is a public private-partnership between NASA Ames Research Center and the SETI Institute with international scope, including private sector engagement from partners including Nvidia, Lockheed Martin, HPE, Google Cloud, Intel, IBM, KX, Canadian Space Agency and Luxembourg Space Agency. Four prior FDL sessions have demonstrated that meaningful progress could be industrialized by bringing together individuals at the PhD and Post Doc level as well as members from industry together to work on connected, but adjacent problems in a shared space mentored by senior scientists with a deep knowledge of the problems. FDL uses sprint methodologies championed by Silicon Valley start-up culture for faster results; places emphasis on interdisciplinary teams for better research quality and viability, and public-private partnerships to lower the costs of high-risk innovation within an expensive technology domain. FDL results will be shared that demonstrate the power of bridging research disciplines and the potential that AI/ML has for supporting research goals, improving on current methodologies, enabling new discoveries, while doing so in accelerated timeframes.

Date and time: 
Tuesday, October 8, 2019 - 2:00pm to 3:00pm