Education

Ph.D. Electrical Engineering and Computer Science, Massachusetts Institute of Technology (Expected Spring 2022)
    S.M. Electrical Engineering and Computer Science, Massachusetts Institute of Technology (2015)
    • Irwin Mark Jacobs and Joan Klein Jacobs Presidential Fellowship (2013)
    B.S. Computer Science and Engineering, The Ohio State University (2012)
    Graduated Summa Cum Laude with Honors Research Distinction in Electrical and Computer Engineering
    • Undergraduate Research Scholarship (2012)
    • Warren G. and James M. Elliott Engineering Scholarship (2011 & 2012)
    • Crowe Horwath Scholarship (2011)
    • TechTomorrow Scholarship (2010)

    Research

    My current research interests span computer vision, interpretable machine learning, and Bayesian non-parametric models. My recent work explores the intersection of probabilistic graphical models and deep learning methods.


    Industry Experience

    Apple Inc., Cupertino, CA
    • Siri, Intern (Summer 2017)
    Google Inc., Mountain View, CA
    • Google Photos, Software Engineering Intern (Summer 2016)
    Systems & Technology Research, Woburn, MA
    • Control and Estimation Systems Group, Intern (Summer 2015)
    MIT Lincoln Laboratory, Lexington, MA
    • Surveillance Systems Group, Co-op (Spring & Summer 2013)
    • Embedded and Open Systems Group, Intern (Summer 2011)

    List of Publications

    Lightweight Data Fusion with Conjugate Mappings

    Christopher L. Dean, Stephen J. Lee, Jason Pacheco, John W. Fisher III
    Preprint: arXiv:2011.10607 (2020)

    We present an approach to data fusion that combines the interpretability of structured probabilistic graphical models with the flexibility of neural networks. The proposed method, lightweight data fusion (LDF), emphasizes posterior analysis over latent variables using two types of information: primary data, which are well-characterized but with limited availability, and auxiliary data, readily available but lacking a well-characterized statistical relationship to the latent quantity of interest. We demonstrate the LDF methodology on two challenging inference problems: (1) learning electrification rates in Rwanda from satellite imagery, high-level grid infrastructure, and other sources; and (2) inferring county-level homicide rates in the USA by integrating socio-economic data using a mixture model of multiple conjugate mappings.

    Efficient MCMC Inference for Remote Sensing of Emission Sources

    S.M. Thesis, Massachusetts Institute of Technology (2015)

    This thesis presents and analyzes an non-parametric MCMC-based inference procedure for identifying the number and properties of gaseous emission sources from measurements taken downwind of the emitters, subject to a specific atmospheric dispersion model. Advised by John W. Fisher III.

    The Value of Delayed Information in Tracking with Distributed Sensor Networks

    B.S. Honors Thesis, The Ohio State University (2012)

    This thesis considers the value of "delayed" observations for object tracking with a variant of the extended Kalman filter where at each time-step a network of communication-constrained sensors can only integrate either one current measurement or one old measurement. Advised by Emre Ertin.

    Student Perspectives on Learning Through Developing Software for the Real World

    Christopher L. Dean, Thomas D. Lynch, Rajiv Ramnath
    Proceedings of 2011 IEEE Frontiers in Education Conference

    We examine several sources of experiential learning for their ability to bridge the gap between the computer science classroom and professional practice. In particular, we compare internships and co-ops, senior capstone projects, and participation in long-term, real-world, student-led software development projects.