NASA Goddard Workshop on Artificial Intelligence
Sponsored by Goddard Chief Technologist
November 27 - 29, 2018
NASA Goddard Space Flight Center - Building 8 Auditorium
AGENDA NOW AVAILABLE (see below)!
For NASA Goddard Employees:
- Room is available in Building 8, Auditorium for those who would like to attend the workshop and poster session.
The Workshop will be live-streamed at all locations of NASA Goddard at the following link:
It will also be recorded and available on the website a few weeks after the end of the Workshop.
Artificial Intelligence (AI) is a collection of advanced technologies that allows machines to think and act, both humanly and rationally, through sensing, comprehending, acting and learning. AI's foundations lie at the intersection of several traditional fields - Philosophy, Mathematics, Economics, Neuroscience, Psychology and Computer Science. Although the inception of AI started in the 1950's, it has recently made a strong comeback in all aspects of society and all over the world; this is mainly due to the timely combination of increased data volumes, advanced and mature algorithms, and improvements in computing power and storage. Current AI applications include big data analytics, robotics, intelligent sensing, assisted decision making, and speech recognition just to name a few.
As stated in the latest NSF Statement on AI for American Industry, "The effects of AI will be profound. To stay competitive, all companies will, to some extent, have to become AI companies." Compared to Industry and Academia, NASA and Goddard have specific challenges as well as resources that are particularly adapted to the use of AI:
- A wealth of data and information to leverage and "learn" from
- Many science- and mission-oriented applications that can benefit from learning on previous data and from domain and expert knowledge
This workshop will be investigating how AI technologies can be adapted or developed to address the following challenges:
- Discover events of interest and correlations in large amounts of science data
- Improve the outcomes of science modeling and data assimilation using improved data processing, integration, and analysis
- Design advisors for mission planning and operations, including anomaly detection and spacecraft health monitoring
- Develop tools for engineering support, including advanced manufacturing, orbit determination, new component design and system engineering
- Customize intelligent user interfaces, including visual analytics and natural language processing
The workshop will be organized as a combination of keynote addresses and short talks followed by focused discussions on specific topics. The areas of interest are as follows:
- General Machine Learning, beyond Neural Networks (NN) techniques and their potential applications to the NASA challenges identified above
- NN and Deep Learning (DL) techniques for supervised and unsupervised learning, as well as specific ML techniques such as NN, including but not limited to Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), and more specifically DL
- Image Data Understanding of remotely sensed imagery, in particular for Onboard Science Data Analysis
- Computer Vision and Image Processing, e.g., for robotic satellite servicing and for extracting and analyzing visual information from documents into higher level information
- Intelligent Advisors for NASA Science and Engineering Applications, including Natural Language Processing for providing easier interfaces to complex systems and that will augment or replace simple web interfaces
- Data Analytics, including Data Mining and Pattern Recognition for Science applications and with special emphasis on:
- Quantification of uncertainty in inference from big data
- Experiment design to create data that is AI/ML ready and robust against misleading correlations
- Methods for prediction of new discovery spaces
- Strength of evidence and reproducibility in inference from big data
- Explainable Artificial Intelligence (XAI). XAI models that provide useful and relevant information to NASA scientists and engineers will be essential to engender acceptance, improve reliability, and develop trust in the systems being developed
|June 1, 2018||Call for Abstracts Release|
|July 31, 2018||Abstract Submission Deadline|
|August 15, 2018 (Target)||Abstract Acceptance Notification|
|August 25, 2018||Workshop Schedule Publication|
|October 31, 2018||Presentation Publication Authorizations Deadline|
|November 27-29, 2018||Workshop Dates|
|January 31, 2019||Presentations Posted on Website|
|February 31, 2019 (Target)||White Paper Summarizing the Workshop|
- Please submit a 500-word text abstract for technical review purposes, that is also suitable for publication. We will circulate your abstract to the workshop technical committee members for review and selection purposes, and then publish the abstract up to 2 weeks prior to the meeting
- Abstracts should contain enough detail to clearly convey the approach and the results of the research
- Commercial papers, papers with no research/development content, and papers where supporting data or a technical description cannot be given for proprietary reasons will not be accepted for presentation in this workshop
- Indicate which of the areas of interest (1 through 7) your abstract most appropriately corresponds to
- Indicate the preference for a short oral presentation (10 minutes) or a poster (although this will be ultimately the technical committee decision)
There is no registration fee. However, the workshop venue has limited space, so pre-registration is mandatory for every attendant (whether presenting an abstract or not). When the maximum number of attendees will have been reached, additional registrations will be placed on a waiting list. Please let us know if you have registered and then decide not to attend.
Additionally, non-US citizens will need to register at least 8 weeks before the workshop, i.e., by August 1st, 2018 at the latest, in order for badges to be processed in time for the workshop.
NASA civil servant and contractor employees are also expected to register in the NASA Conference Tracking System (NCTS) to attend the workshop, even for non-cost, local attendance. Visit https://ncts.nasa.gov to complete this process. The NCTS number for the NASA Goddard Workshop on Artificial Intelligence is 36893-19.