Esther Rolf Personal Website

I am an assistant professor of Computer Science at the University of Colorado, Boulder, the AI/ML lead for the Environmental Data Science Innovation and Impact Lab (ESIIL) and a scientific director for the Togo Data Lab.

I research statistical and geospatial machine learning. My research blends methodological and applied techniques to study and design machine learning algorithms and systems with an emphasis on usability, data-efficiency and fairness. My current research directions include developing algorithms and infrastructure for reliable environmental monitoring using machine learning, and understanding the multifaceted nature of representation in data and how that affects our ability to train fair and effective machine learning systems. My work has been recognized with an SDG Digital Gamechangers award (2023) from the United Nations Development Programme and the International Telecommunication Union, and best paper awards at ICML (2018), the Workshop on AI for Social Good at Neurips (2019), and the Workshop on Machine Learning for Remote Sensing at ICLR (2025).

For an overview of my research, you can watch my keynote talk from the NeurIPS workshop on Tackling Climate Change with Machine Learning from 2024.

Research

I develop and analyze of machine learning algorithms to address pressing social and environmental problems. Sometimes this entails developing or analyzing new statistical ML techniques, sometimes this entails carefully applying, adapting, and evaluating ML methods in a specific context of use; most often it entails a blend of the two. Some specific project areas and application contexts:

Earth Embeddings: high dimensional embeddings of Earth and its data; achieving this with implicit neural representations (e.g., SatCLIP).
Tailored machine learning for remotely sensed data (e.g. using satellite imagery for environmental monitoring).
Characterizing and formalizing notions of representativeness in training data (e.g. numerical representation: how many data points come from each source or group?, as distinct from what components of an individual or environment does a collection of data actually reflect?), and how these notions of representation affect our ability to train fair and useful machine learning systems.
Understanding and addressing key challenges in geospatial machine learning (e.g. spatial error structures make it hard to evaluate geospatial ML models, and can introduce concerns of bias or unfairness in downstream use).

For a full list of papers please see my google scholar page.

Research Group

I am fortunate to get to work with incredible students, postdocs, and researchers. Members of the research group are doing innovative (sometimes interdisciplinary) research, and are expected to contribute to a collaborative and supportive group culture.

PhD Students

Livia Betti (2024-), CU Chancellor's Fellow and NSF GRFP Fellow

Postdocs

Levi "VeeVee" Cai (2025-), ESIIL AI Postdoctoral Fellow

Master's students

Ian Christie
Lucia Wittiko

Research mentees

Ruth Crasto

Interested in a PhD, postdoc, or other involvment with the lab? I am recruiting PhD Students to join my lab at CU Boulder. Please first read my guide to getting involved with the lab, which includes instructions for the best way to contact me. CU-specific resources are linked below. Please note that I will not be able to respond to all emails.

Courses

Fall 2024, Fall 2025: Current Topics in Computer Science: Geospatial and Statistical Machine Learning

Spring 2025, Spring 2026: (Graduate) Machine Learning

Course syllabus (2025)

Resources

PhD application resources

MOSAIKS: Generalizable and accessible machine learning with global satellite imagery

The MOSAIKS data interface has moved to api.mosaiks.org!
See the MOSAIKS project page for more resources and updates.