Sebastian Pokutta

David M. McKenney Family Associate Professor

Associate Director of the Center for Machine Learning @ GT


Georgia Institute of Technology

Industrial and Systems Engineering (ISyE)

Algorithms and Randomness Center (ARC)

Center for Machine Learning @ GT (ML@GT)


Email. <firstname.lastname> AT

Twitter. @spokutta

Prospective Students/Postdocs. Please note that I unfortunately will be unable to respond to most inquiries regarding openings in my group. The admissions process at Georgia Tech is handled by the respective departments and not individual faculty. If you are already a student at GT or have been admitted, please feel free to contact me if you are interested in my research.

Recent Papers.

  • Restarting Frank-Wolfe
    T. Kerdreux, A. d'Aspremont, S. Pokutta
    preprint (2018), [arXiv]
  • Principled Deep Neural Network Training through Linear Programming
    D. Bienstock, G. Muñoz, S. Pokutta
    preprint (2018), [arXiv]
  • Limits of Treewidth-based tractability in Optimization
    Y. Faenza, G. Muñoz, S. Pokutta
    preprint (2018), [arXiv] [summary]
  • Blended Conditional Gradients: the unconditioning of conditional gradients
    G. Braun, S. Pokutta, D. Tu, S. Wright
    preprint (2018), [arXiv]

Select Recent Talks.

  • 05/2018: (technical) "Blended Conditional Gradients" at Monash University [slides] (unfortunately the animations do not work; see [arXiv] for exact algorithm)
  • 04/2018: (non-technical) "When the AI God fails: Recent Adventures in AI Hacking, Security, and Fairness" at the Hong Kong Analytics Community Meeting [slides]
  • 04/2018: (technical) "Emulating the Expert: Inverse Optimization through Online Learning" at the "Optimization and Discrete Geometry" Workshop in Tel Aviv [slides]