Florian Schäfer

331 Annenberg

1200 E California Blvd

Pasadena, CA 91106

I’m Florian Schäfer, PhD-candidate in applied and computational mathematics at Caltech, where I’m doing research at the interface of partial differential equations, statistical inference, and fast algorithms. More recently, I have been working on game theory as a paradigm for algorithm design and the computational challenges that come with it. Check out these high-level summaries if you want to know more!

I’m part of the Department of Computing and Mathematical Sciences and my PhD-advisor is Houman Owhadi.

Before coming to Caltech, I obtained my Bachelor’s– and Master’s degrees in Mathematics at the University of Bonn.

I am on the academic job market this fall (2020/2021). My teaching and research statements are available upon request.

news

Sep 23, 2020 Check out my talk at the Second Symposium on Machine Learning and Dynamical Systems at the Fields institute.
Sep 17, 2020 The ICML 2020 talks are now online! Check out my talk on implicit competitive regularization at the main conference and my talk on competitive mirror descent at the workshop “Beyond first order methods in machine learning systems”.
Aug 11, 2020 Check out my talk at the Bernoulli-IMS One World Symposium 2020. I am talking about joint work with Matthias and Houman on the sparse factorization of dense kernel matrices using Kullback-Leibler minimization.
Jul 31, 2020 I will give a virtual talk at Mila on August 5th, 2:30pm Eastern
Jul 31, 2020 Check out the new blog post on our work on sparse Cholesky factorization of dense kernel matrices.

selected publications

  1. Sparse Cholesky factorization by Kullback-Leibler minimization
    Schäfer, Florian, Katzfuss, Matthias, and Owhadi, Houman
    arXiv Talk
  2. Compression, inversion, and approximate PCA of dense kernel matrices at near-linear computational complexity
    Schäfer, Florian, Sullivan, T. J., and Owhadi, Houman
    To appear in SIAM Multiscale Modeling and Simulation
    arXiv Blog Talk
  3. Implicit competitive regularization in GANs
    Schäfer, Florian, Zheng, Hongkai, and Anandkumar, Anima
    In the 37th International Conference on Machine Learning (ICML 2020)
    arXiv Blog Slides Talk
  4. Competitive Gradient Descent
    Schäfer, Florian, and Anandkumar, Anima
    In Advances in Neural Information Processing Systems 32
    arXiv HTML Blog Poster Slides