Thomas Flynn

Research Interests

My research interests and recent projects include:

  • Provable optimization methods for machine learning
  • Gradient estimation for Markov chains
  • Change detection and its applications

I am very interested in first order gradient methods, particularly in the context of non-linear machine learning problems. I am also interested in the role of geometry in numerical computing, including optimization on manifolds and contraction methods. Basically, anything with a sufficiently mathematical flavor, at the intersection of optimization, machine learning, probability, etc.. If any of these topics also interest you then feel free to get in touch!

Brief Bio

Since January 2018 I am a postdoctoral research associate in the Computational Science Center at Brookhaven National Lab. My supervisor is Shinjae Yoo

I received my PhD in Computer Science from the CUNY Graduate Center. During my studies I was member of the CoSSMO Lab directed by Prof. Felisa Vázquez-Abad.

Click here to view my CV.


Office: Hunter North 1000C (CoSSMO Lab)

Preprints and Publications


  • manseg is a program for manual labeling of 3D point cloud data. Think MS paint for point clouds. bitbucket page