For more information about this meeting, contact Jason Morton.

Title: | Geometry of maximum likelihood estimation in Gaussian graphical models |

Seminar: | Applied Algebra and Network Theory Seminar |

Speaker: | Caroline Uhler, U.C. Berkeley |

Abstract: |

We study maximum likelihood estimation in Gaussian graphical models from a geometric point of view. In current applications of statistics, we are often faced with problems involving a large number of random variables, but only a small number of observations. An algebraic elimination criterion allows us to ﬁnd exact lower bounds on the number of observations needed to ensure that the maximum likelihood estimator exists with probability one. This is applied to bipartite graphs and grids, leading to the ﬁrst instance of a graph for which the maximum likelihood estimator exists with probability one even when the number of observations equals the treewidth of the graph. |

### Room Reservation Information

Room Number: | MB106 |

Date: | 09 / 14 / 2011 |

Time: | 02:30pm - 03:20pm |