CV

Ph.D.

University of New Mexico, School of Engineering

Computer Science

M.B.A.

University of New Mexico, Anderson School of Business

Business Management

B.S.

University of New Mexico, School of Engineering

Computer Science

Postdoctoral Appointee

Sandia National Laboratories

Postdoc in the Scientific Machine Learning department.

R&D Graduate Intern

Sandia National Laboratories

Year-round intern in the Scientific Machine Learning department.

• Participated in the Lab Driven Research & Development project CLDERA, which seeks to develop tools to identify source to impact pathways in climate systems.
• Researched the application of causal discovery, specifically the PC and PCMCI algorithms, for identifying causal pathways from the 1991 Mt. Pinatubo eruption.
• Developed the CaStLe algorithm for constructing spatiotemporal causal graphs of local dependencies in climate data.
• Researched random forest machine learning techniques for learning about emergent dynamics in climate models.
• Used feature importance analysis to make comparisons between observed and simulated data to look for the simulations' faults

Graduate Research Assistant

University of New Mexico

A research assistant under Dr. Lydia Tapia and Dr. Marina Kogan.

• Robotics under Dr. Tapia
• Computational sociology under Dr. Kogan

Owner

Swarming Technologies LLC

Ran a business full-time manufacturing, repairing, and developing autonomous ground-based robots.

• Robots, known as 'Swarmies' were designed for swarm robotics and autonomous robotics research and education.
• Swarmies were featured in the NASA Swarmathon, NASA Minds competition, and CS4All NM.

Robotics Engineer and Designer

NASA Swarmathon & Moses Biological Computation Lab

Designed and developed autonomous, swarming robots called Swarmies. The robots were a part of a nation-wide, college-level robotics competition called the NASA Swarmathon. Development included designing with Autodesk Inventor and manufacturing parts with SLS, SLA, and FDM 3D printing, as well as electronics development.

• Analyzed data to track progress and success determinants using MySQL and Python’s numpy, scipy, and pandas.
• Developed automated deployment for code on robots using Ansible and Docker.
• Conducted swarm robotics research using Arduino/iPod Touch-controlled robots, iAnts, using a genetic algorithm to tune behavior.

Software Engineering Intern

Intel Corporation

Product surveying, QA, troubleshooting, and testing.

• Arduino/Galileo programming and debugging.
• Design and construction of a robotic car, controlled by the Intel Galileo Gen 2.
• Developed a Arduino/Intel Galileo controlled robot that included: input given via Bluetooth from Android phone and multiple sensors for line following and obstacle avoidance.

Best Talk Prize

European Seminar on Computing (ESCO)

3rd Place Poster Prize

Department of Energy Conference on Data Analysis (CoDA)

Adaptive Ski Instructor

Adaptive Sports Program New Mexico Inc

Teach skiing to people with various disabilities.

Troop 9 Board of Review Member

Boy Scouts of America

Attend board of review meetings to assist in scout advancement.

Space-Time Causal Discovery in Earth System Science: A Local Stencil Learning Approach

Journal of Geophysical Research: Machine Learning and Computation

Causal discovery tools enable scientists to infer meaningful relationships from observational data, spurring advances in fields as diverse as biology, economics, and climate science. Despite these successes, the application of causal discovery to space-time systems remains immensely challenging due to the high-dimensional nature of the data. For example, in climate sciences, modern observational temperature records over the past few decades regularly measure thousands of locations around the globe. To address these challenges, we introduce Causal Space-Time Stencil Learning (CaStLe), a novel meta-algorithm for discovering causal structures in complex space-time systems. CaStLe leverages regularities in local space-time dependencies to learn governing global dynamics. This local perspective eliminates spurious confounding and drastically reduces sample complexity, making space-time causal discovery practical and effective.

Benchmarking the PCMCI Causal Discovery Algorithm for Spatiotemporal Systems

OSTI

Causal discovery algorithms construct hypothesized causal graphs that depict causal dependencies among variables in observational data. While powerful, the accuracy of these algorithms is highly sensitive to the underlying dynamics of the system in ways that have not been fully characterized in the literature. In this report, we benchmark the PCMCI causal discovery algorithm in its application to gridded spatiotemporal systems. Effectively computing grid-level causal graphs on large grids will enable analysis of the causal impacts of transient and mobile spatial phenomena in large systems, such as the Earth's climate.

Learning Why: Data-Driven Causal Evaluations of Climate Models.

ICML 2021 Workshop Tackling Climate Change with Machine Learning

We plan to use nascent data-driven causal discovery methods to find and compare causal relationships in observed data and climate model output. We will look at ten different features in the Arctic climate collected from public databases and from the Energy Exascale Earth System Model (E3SM). In identifying and comparing the resulting causal networks, we hope to find important differences between observed causal relationships and those in climate models. With these, climate modeling experts will be able to improve the coupling and parameterization of E3SM and other climate models.

Causal Evaluations for Identifying Differences between Observations and Earth System Models

OSTI

Machine learning feature analysis illuminates disparity between E3SM climate models and observed climate change

Journal of Computational and Applied Mathematics

In September of 2020, Arctic sea ice extent was the second-lowest on record. State of the art climate prediction uses Earth system models (ESMs), driven by systems of differential equations representing the laws of physics. Previously, these models have tended to underestimate Arctic sea ice loss. The issue is grave because accurate modeling is critical for economic, ecological, and geopolitical planning. We use machine learning techniques, including random forest regression and Gini importance, to show that the Energy Exascale Earth System Model (E3SM) relies too heavily on just one of the ten chosen climatological quantities to predict September sea ice averages. Furthermore, E3SM gives too much importance to six of those quantities when compared to observed data. Identifying the features that climate models incorrectly rely on should allow climatologists to improve prediction accuracy.

Arctic Tipping Points Triggering Global Change (LDRD Final Report)

OSTI

The Swarmathon: An Autonomous Swarm Robotics Competition

arXiv

The Swarmathon is a swarm robotics programming challenge that engages college students from minority-serving institutions in NASA's Journey to Mars. Teams compete by programming a group of robots to search for, pick up, and drop off resources in a collection zone. The Swarmathon produces prototypes for robot swarms that would collect resources on the surface of Mars. Robots operate completely autonomously with no global map, and each team's algorithm must be sufficiently flexible to effectively find resources from a variety of unknown distributions.