Photo

Shahrouz Ryan Alimo, Ph.D.

sralimo@jpl.nasa.gov

         



Ryan Alimo is a machine learning scientist and intrapreneur at the Jet Propulsion Laboratory, California Institute of Technology's Deep Learning Group. Ryan Alimo's research interests span theory and practice of black-box optimization for machine learning. In particular, he has been working on the development and analysis of machine learning algorithms for teams of spacecraft, which can cooperate to form virtual structures such as synthetic apertures radar (SAR), and can perform distributed measurements that is usually not possible with a single spacecraft.

2018 - present

Data scientist and project leader

Mission Control Systems Deep Learning Technology Group

NASA's Jet Propulsion Laboratory

2017 - 2018

Postdoc in Caltech's Center for Autonomous Systems and technology

Caltech's CAST

2014 - 2017

Ph. D. in Computational Science

Efficient data-driven global optimization approaches, with applications in computer-aided designs

Advisor: Thomas R. Bewley

UC San Diego

Summer 2015

Internship in uncertainty quantification and machine learning

Fujitsu Lab of America

2012 - 2014

M. Sc. in Mechanical Eng.

Specialization in High Performance Computing

Advisor: Alison Marsden

UC San Diego

2007 - 2012

B. Sc. in Mechanical Eng.

Using GPU for numerical computation via CUDA

Advisor: Mir A. Jalali

B.Sc. Sharif University of Technology

2005

Gold Medal of Astronomy Olympiad

Iranian National Astronomy Olympiad

Awards

  • Voyager Award 2019, NASA's JPL
    • (In recognition for the development of innovative optimization algorithms for deep learning techniques to accelerate automation in mission operation systems)
  • Outstanding Graduate Student Award 2017, University of California San Diego (UCSD)
    • (The successful candidate must show strong research initiative and leadership skills)
  • ‘Sharif Stars’ Prize for outstanding performance at Sharif University of Technology, 2009
  • Gold Medal (1st rank among 100,000 students) Iranian National Astronomy Olympiad, 2005

People



I had the pleasure of mentoring and working with several talented students at:



NASA's JPL,

  • Sharmita Dey (Ph.D. student at University of Goettingen), 2021-now
  • Thomas Claudet (M.Sc. student at Georgia Tech), 2020-now
  • Jonathan Alex Sabol (Ph.D. student at University of Texas, Arlington), 2019-now
  • Dounia Lakhmiri (Ph.D. student at Polytechnique Montréal), 2019-now
  • Tommaso Guffanti (Ph.D. student at Stanford University), 2019-20
  • Antoni Viros-i-Martin (Ph.D. student at Texas A&M University), 2019-20
  • Dr. Kingson Man (Ph.D. and M.Sc. at USC), 2019-20
  • Kasra Yazdani (Ph.D. student at University of Florida), summer 2019
  • Daniel Jeong (B.Sc. Columbia University), summer 2019
  • Jonathan Chu (B.Sc. Harvard University), summer 2019
  • Dylan Sam (B.Sc. Brown University), summer 2019
  • Michael Lally (B.Sc. University of Illinois at Urbana-Champaign), summer 2019
  • Alex Gao (B.Sc. University of California, Berkeley), summer 2019
  • Hunter Hall (B.Sc. University of California, Berkeley), summer 2019
  • Christine Yuan (B.Sc. Cornell University), summer 2019
  • Luan Nguyen (B.Sc. UC San Diego), summer 2019 ), summer 2019
  • Jonathan Margoliash (M.Sc. at UC San Diego), summer 2018
  • Andrew Hess (B.Sc. at Caltech), summer 2018


UC San Diego, and Caltech:

  • Muhan Zhao (Ph.D. Student at UC San Diego), 2016-18
  • Abhishek Subramanian (M.Sc. UCSD, now a Product Manager at Perma Pure LLC), 2016-17
  • Karena Kai (Ph.D. Candidate at Caltech), Fall 2017
  • Thayjes Srivas (M.Sc. Research Engineer Caltech), 2018
  • Karthik Karnik (B.Sc. at Caltech), Spring 2018
  • Kapil Sinha (B.Sc. at Caltech), Spring 2018

Publications

  • Alimo, S.R., Beyhaghi, P. and Bewley, T.R., 2019. Delaunay-Based Global Optimization in Nonconvex Domains Defined by Hidden Constraints. In Evolutionary and Deterministic Methods for Design Optimization and Control With Applications to Industrial and Societal Problems (pp. 261-271). Springer, Cham.
  • V Capuano, SR Alimo, AQ Ho, SJ Chung,Robust Features Extraction for On-board Monocular-based Spacecraft Pose Acquisition,AIAA Scitech 2019 Forum, 2019
  • E Bıyık, J Margoliash, SR Alimo, D Sadigh, Efficient and Safe Exploration in Deterministic Markov Decision Processes with Unknown Transition Models, arXiv preprint arXiv:1904.01068, 2019
  • M Zhao, SR Alimo, TR Bewley, An active subspace method for accelerating convergence in Delaunay-based optimization via dimension reduction, 2018 IEEE Conference on Decision and Control (CDC), 2018
  • Beyhaghi, P., Alimo, S. R., Bewley, T.,A derivative-free optimization algorithm for the efficient minimization of functions obtained via statistical averagings (submitted to Journal of Computational Optimization and Applications)
  • Alimo, S. R., Beyhaghi, P., Bewley, T.,Delaunay-based derivative-free optimization via global surrogates: nonconvex constraints (submitted to Journal of Global Optimization)
  • Alimo, S. R., Beyhaghi, P., Cavaglieri, D., Bewley, T.,Design of IMEXRK time integration schemes via Delaunay-based derivative-free optimization with nonconvex constraints and grid-based acceleration (to submit for Journal of Global Optimization)
  • Alimo, S. R., Beyhaghi, P., Bewley, T.,Implementation of dense lattices to accelerate Delaunay-based optimization (to submit for Journal of Optimization and Engineering)
  • Beyhaghi, P., Alimo, S. R., Bewley, T. R.,A multiscale, asymptotically unbiased approach to uncertainty quantification in the numerical approximation of infinite time-averaged statistics, arXiv preprint arXiv:1802.01056
  • Alimo, S. R., Beyhaghi, P., Bewley, T. R.,Optimization combining derivative-free global exploration with derivative-based local refinement (submitted to Control Decision Conference, CDC-2017)
  • Alimo, S. R., Beyhaghi, P., Bewley, T. R.,Delaunay-based global optimization in domains defined by hidden constraints(accepted to Surrogate-assisted Optimization at EUROGEN 2017)
  • Alimo, S., Beyhaghi, P., Meneghello, G., Bewley, T.,Delaunay-based optimization in CFD leveraging multivariate adaptive polyharmonic splines (MAPS) (published in AIAA SciTech 2017)
  • Beyhaghi, P., Alimo, S. R., and Bewley, T. R.,Delaunay-based method for optimizing infinite time averages of numerical discretizations of ergodic systems (under prepration)
  • Subramanian, A., Alimo, S. R., Gill, P. E., Bewley, T., Optimizing path for an object finding robot (under preparation)
  • Alimo, S., He, D., Multi-stage algorithm for uncertainty quantification of solar power forecasts (published in IEEE power system conference 2016)
  • He, D., Alimo, S. Operating a solar power generating system, U.S. patent, US 20170117711 A1