Machine Learning and Instrument Autonomy Group

     Peyman Tavallali is a science data machine learning researcher and data scientist at NASA’s Jet Propulsion Laboratory (JPL). Peyman is doing research on a diverse set of problems including the Ocean Worlds Life Surveyor (OWLS), Anomalous vs Nominal Identification & Monitoring of Underlying States (ANIMUS), and Unified Processing for Robotic Icy Terrain Exploration (UPRITE). In these projects, Peyman is using and designing algorithms in the fields of tracking, anomaly detection, streams clustering, and reinforcement learning.

     Before joining JPL, Peyman was a co-founder and Chief Information Officer (CIO) at Avicena LLC, a start-up company in biomedical engineering and data science.

    

Contact

Email: peyman.tavallali@jpl.nasa.gov
Office: 158-252E

Mail Stop: 158-242
4800 Oak Grove Drive
Pasadena, CA 91109

Education

Applied and Computational Math (PhD) - California Institute of Technology

Research Interests

Machine Learning
Data Clustering
Tracking
Decision Trees
Reinforcement Learning
Artificial Intelligence
Deep Neural Networks
Biomedical Signal Processing and Machine Learning
Adaptive Signal Processing
Mathematical Modeling
Optimization
Numerical Analysis
Numerical and Analytic solutions of Differential Equations
Scientific Programming

Publications

Peyman Tavallali, Steve Chien, Lukas Mandrake, Yuliya Marchetti, Hui Su, Longtao Wu, Benjamin Smith, Andrew Branch, James Mason, and Jason Swope. Adaptive, Model-driven Observation for Earth Science. 101st American Meteorological Society Annual Meeting, 2021.

Michael C. Stanley, Steffen Mauceri, Helene L Seroussi, Peyman Tavallali, Hamed Hamze Bajgiran, and Houman Owhadi. Uncertainty estimation of basal friction using a game theoretical approach. AGU Fall Meeting 2020.

Hui Su, Longtao Wu, Jonathan H Jiang, Raksha Pai, Alex Liu, Albert J Zhai, Peyman Tavallali, and Mark DeMaria. Applying satellite observations of tropical cyclone internal structures to rapid intensification forecast with machine learning. Geophysical Research Letters, 2020.