Sarah Marzen is a distinguished physicist and interdisciplinary researcher whose work bridges information theory, cognitive science, and biology. As an associate professor, she has contributed extensively to the study of sensory prediction, reinforcement learning, and resource rationality, securing leadership roles in numerous federally funded research projects. Her academic background includes a Ph.D. from the University of California, Berkeley, and postdoctoral work at MIT. She has published widely in peer-reviewed journals and played a vital role as a guest editor for multiple special issues. Sarah is actively involved in professional service, mentoring, and organizing scientific workshops. Her research stands out for its originality and interdisciplinary reach, tackling complex questions in neural computation and theoretical biology. Through her editorial work, teaching, and committee service, she has helped shape the scientific community’s understanding of cognition and prediction. Sarah Marzen’s scholarly excellence and leadership position her as a significant figure in contemporary scientific research.

Professional Profile 

Google Scholar | Scopus Profile

Education

Sarah Marzen pursued her undergraduate studies in physics at the California Institute of Technology, where she developed a strong foundation in theoretical and experimental research. She continued her academic journey at the University of California, Berkeley, earning a Ph.D. in physics. Her doctoral work focused on bio-inspired problems in rate-distortion theory, under the guidance of Professor Michael R. DeWeese. This research bridged information theory and biological systems, laying the groundwork for her future interdisciplinary pursuits. In addition to her formal degrees, she attended several prestigious summer schools and workshops, including the Santa Fe Institute’s Complex Systems School and the Machine Learning Summer School. These programs helped her expand her understanding of machine learning, complex systems, and computational neuroscience. Sarah’s educational background is marked by both academic excellence and a consistent interest in the convergence of physics, information theory, and biological intelligence, making her uniquely equipped for innovative cross-disciplinary research.

Experience

Sarah Marzen’s academic career reflects deep engagement with both research and teaching. She currently serves as an associate professor of physics at the W. M. Keck Science Department, affiliated with Claremont McKenna, Pitzer, and Scripps Colleges. Prior to this, she was an assistant professor in the same department and a postdoctoral fellow at MIT, where she worked with Professors Nikta Fakhri and Jeremy England. Her early research experience includes graduate work at UC Berkeley and multiple assistantships and fellowships during her undergraduate years at Caltech. She has also held advisory roles in academia and private research, such as mentoring for Google Summer of Code and advising a stealth startup. Her experience spans experimental physics, theoretical modeling, machine learning, and neuroscience. Alongside her teaching, she contributes significantly to committee service and program development within her department, reflecting a well-rounded academic profile. Her professional trajectory demonstrates a strong commitment to both discovery and mentorship.

Research Focus 

Sarah Marzen’s research centers on understanding how intelligent systems—both biological and artificial—predict and adapt to their environments. Her primary focus areas include sensory prediction, reinforcement learning, and resource rationality, particularly through the lens of information theory. She explores the ways in which brains and machines can perform efficient, predictive computations under constraints, contributing to theoretical frameworks that bridge physics, neuroscience, and cognitive science. Her work has applications in neural networks, artificial intelligence, and computational biology. She also investigates how delayed feedback and memory structures affect learning dynamics, as reflected in her studies of reservoir computing and time-delayed decision processes. Through her interdisciplinary approach, she addresses fundamental questions about how information is processed and used by complex systems. Her research aims to uncover principles of learning and adaptation that apply across different domains of intelligence, providing insight into both natural cognition and the design of intelligent machines.

Award and Honor