Image: U.S. Marine Corps Lt. Col. Pedro Ortiz, a computer science doctoral student at the Naval Postgraduate School (NPS), works with his advisor, Assistant Professor Marco Oleskanin, to enable fast and effective decision-making by Navy commanders. My doctoral thesis focused on making An era of ever-increasing sensor data and uncertainty.
opinion more
Credit: U.S. Navy Photo Credit: Javier Chagoya
The proliferation of battlefield sensors and increased data makes it easy for commanders to fall into the information paradox. So he’s drowning in data and hungry for knowledge.
U.S. Marine Corps Lieutenant Colonel Pedro Ortiz graduated from the Naval Postgraduate School (NPS) on June 16 with a Ph.D. With his Ph.D. in Computer Science, he has applied his expertise to this challenge to enable quick and effective decision-making by commanders in an era of ever-increasing sensor data and uncertainty. PhD thesis focused.
“I am very interested in applying artificial intelligence and machine learning to solve combatant problems,” Ortiz said. “Ideally, I would like to be able to research a combat problem, come up with some solutions, and lead efforts to apply those solutions to the operational force.”
Ortiz’s doctoral dissertation titled “Quantification and Decomposition of Uncertainty with Bayesian Deep Learning for Big Data Satellite Remote Sensing Problems” was just an exercise in putting it into practice. Ortiz describes uncertainty quantification (UQ), which is crucial in reducing the impact of uncertainty in optimization and decision-making processes through the application of probabilistic models to large satellite remote sensing datasets. I considered the method.
Ortiz joins an elite group of PhDs in the Marine Corps. program (PHDP). This highly competitive program (only two candidates are selected each year) supports senior leadership decision-making, assists in the development of defense and military strategy, and informs long-term concepts and capabilities. We provide services by a group with strategic and advanced technical ideas that support development area.
In his next assignment, Ortiz will report to the Department of Defense’s Office of the Director of Digital and Artificial Intelligence (CDAO) and apply his newfound knowledge and expertise. Ortiz said his research could have a positive impact on joint all-area command and control (JADC2) in current and future operations.
“My research involved using data sets from two different satellites and two different sensors on each of those satellites,” he explained. “We used probabilistic deep learning to fill in the gaps in our microwave dataset using infrared from another satellite. I was able to tell them how much I can trust them.”
UQ methods have been applied to solve a variety of real-world problems in science and engineering.
“With more data to fill the gaps between different sensors and an understanding of data uncertainty, combatants will have more information to make better decisions. We did,” Ortiz said. “High uncertainty means that the model output may not be reliable. Understanding the uncertainty makes the model output more interpretable. and other complex automated military systems.
Ortiz says predictive modeling ultimately helps decision-making because it addresses a key component of reliability in artificial intelligence and machine learning.
“I hope that my research will inspire others to use the same model as I do and to reap the benefits of being able to measure uncertainty,” he continued. “One of the main contributions I have made to him is the use of this additional information in many scientific fields, basically in any field where deep learning is currently used or may be used in the future. I figured out how.”
Ortiz began his Ph.D. It was right around the time that COVID-19 hit the world and NPS transitioned to his 100% online education.
“The PhD, the program was rigorous and really expanded my thinking,” Ortiz said. “Despite the pandemic, there is still a lot of progress in my field and finding ways to make a unique contribution has been as difficult as it was before the pandemic. I had a great adviser, I was able to publish a magazine article in my first year at NPS, which is not always the norm.
“My thoughts have changed over time, too. I wrote my dissertation proposal in the first nine months, but two years later my doctoral dissertation was very different. It’s not that surprising because it’s the way you come up with ideas, test them, adjust them, iterate,” he continued.
A lifelong learner, Ortiz’s professional approach naturally brought her to research, both at the peak of the pandemic and when schools returned to normal. In fact, this is Ortiz’s second educational tour with his NPS. He graduated with his Master of Science in Computer Science in 2010, completing his studies three months early and winning the Distinguished Paper Award. Ortiz said he was pleased to see that NPS now has several master’s students and that he is already working on a follow-on field for his doctorate. research.
Despite the initial challenges, Ortiz reveled in the rigor of his plan and discovered a unique problem directly related to the operationally important issues of the Navy. Ortiz also graduated with the distinction of being the first Hispanic Marine to receive a doctorate degree. Through the Corps’ PHDP-Technical (PHDP-T) program.
Learn more about Ortiz’s research here: https://youtu.be/b-s_i3cGxjY
Disclaimer: AAAS and EurekAlert! We are not responsible for the accuracy of news releases posted on EurekAlert!. Use of information by contributing institutions or via the EurekAlert system.
