Efficient Robust Proper Learning of Log-concave Distributions

  title={Efficient Robust Proper Learning of Log-concave Distributions},
  author={Ilias Diakonikolas and Daniel M. Kane and Alistair Stewart},
We study the robust proper learning of univariate log-concave distributions (over continuous and discrete domains). Given a set of samples drawn from an unknown target distribution, we want to compute a log-concave hypothesis distribution that is as close as possible to the target, in total variation distance. In this work, we give the first computationally efficient algorithm for this learning problem. Our algorithm achieves the information-theoretically optimal sample size (up to a constant… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 11 times over the past 90 days. VIEW TWEETS

From This Paper

Topics from this paper.


Publications citing this paper.


Publications referenced by this paper.
Showing 1-10 of 47 references


  • I. Diakonikolas, D. M. Kane, A. Stewart. Properly learning poisson binomial distri CoRR
  • To appear in COLT
  • 2016
1 Excerpt


  • Q. Han, J. A. Wellner. Approximation, estimation of s-concave densities via renyi divergences. Ann. Statist.
  • 06
  • 2016


  • C. R. Doss, J. A. Wellner. Global rates of convergence of the mle logconcave, s-concave densities. Ann. Statist.
  • 06
  • 2016


  • C. Daskalakis, A. De, G. Kamath, C. Tzamos. A size-free CLT for poisson multinomials, its applications
  • abs/1511.03641, 2015. To appear in STOC
  • 2016

Similar Papers

Loading similar papers…