Private pilot's license (airplane single-engine land), 2016
My background is in Computer Science, Planetary Science, and
Geology. I am most interested in problems that lie at the interfaces
between these fields, such as automated methods (artificial
intelligence, machine learning) to investigate science questions using
planetary data (orbital and in situ).
I work at the Jet
Propulsion Laboratory in Pasadena, CA, as a researcher in the
Machine Learning and Instrument
Autonomy Group, investigating ways that machine learning can be
used to increase the autonomy of space missions.
From 2013 to 2017, I also served as a
tactical planner and uplink lead for the
Mars Exploration Rover Opportunity. Since 2018, I have also served
as the PDS Imaging Node
I am also an Associate Research Professor at Oregon State University,
where I have the pleasure of
teaching classes in Computer Science and doing research on competency-aware machine learning.
My research projects at JPL have included:
- Onboard science for Europa Clipper: Developing and testing methods
to quickly analyze data as it is collected during a flyby of Europa to
assign high downlink priorities to the most scientifically valuable
observations and to enable cross-instrument collaboration
- Mars Target Encyclopedia: Information extraction from scientific
publications for planetary science
- V-FASTR: Efficient machine learning to detect transient radio phenomena (e.g., pulsars and Fast Radio Bursts) in real time
- Collaborative machine learning for sensor networks
- Automatic landmark identification and change detection in Mars orbital images (dark slope streaks, dust devil tracks, etc.)
- Analyzing the sensitivity of machine learning algorithms to high-radiation environments
- Predicting county-level crop yield from Earth orbital images
- Modeling user preferences for sets, rather than individual items
(like music playlists or rover image downlink sets)
- Modeling flight software with state charts and using automatic code generation to convert them into C/C++ (for implementation) or Promela (for model checking)
- Tracking the north polar ice caps (water and CO2) on Mars
News and upcoming events:
- Press release (Oct. 1, 2020):
AI Is Helping Scientists Discover Fresh Craters on Mars
- The video for my controversial ICML 2012 talk is no longer available (lost in a server crash).
However, you can read the original paper:
Machine Learning that Matters (pdf, 6 pages, 234K) and
see the slides from a subsequent invited AAAI talk:
Challenges for Machine Learning Impact on the Real World (1.6M).
- Recently published or posted:
- Novelty-Driven Onboard Targeting for Mars Rovers.
Kiri L. Wagstaff, Raymond Francis, Hannah Kerner, Steven Lu,
Favour Nerrise, James F. Bell III, Gary Doran, and Umaa Rebbapragada.
Proceedings of the International Symposium on Artificial Intelligence, Robotics and Automation in Space, 2020.
Efficient Active Learning in New Domains.
Kiri Wagstaff and Steven Lu.
ICML Workshop on Real World Experiment Design and Active Learning, 2020.
- How can we train classifiers when we don't know what all
possible classes are before we begin? To support
discovery in exploration settings, we employ active learning
to help with discovering classes and labeing data.
We apply this approach to a Mars rover data set.
- Data set (labeled Mars rover images):
- All publications