## Seminar: NIPS previews

Speaker | Stephen Pasteris, Wittawat Jitkrittum, Ricardo Silver |
---|---|

Affiliation | UCL |

Date | Friday, 04 Dec 2015 |

Time | 13:00 - 14:00 |

Location | Roberts G08 (Sir David Davies lecture theatre) |

Event series | DeepMind CSML Seminar Series |

Description |
Talk 1: Stephen Pasteris will present Online Prediction at the Limit of Zero Temperature by Mark Herbster and Stephen Pasteris Abstract: We design an online algorithm to classify the vertices of a graph. Underpinning the algorithm is the probability distribution of an Ising model isomorphic to the graph. Each classification is based on predicting the label with maximum marginal probability in the limit of zero-temperature with respect to the labels and vertices seen so far. Computing these classifications is unfortunately based on a #P- complete problem. This motivates us to develop an algorithm for which we give a sequential guarantee in the online mistake bound framework. Our algorithm is optimal when the graph is a tree matching the prior results in [?]. For a general graph, the algorithm exploits the additional connectivity over a tree to provide a per-cluster bound. The algorithm is efficient, as the cumulative time to sequen- tially predict all of the vertices of the graph is quadratic in the size of the graph. Talk 2: Wittawat Jitkrittum will present Bayesian Manifold Learning: The Locally Linear Latent Variable Model by Mijung Park, Wittawat Jitkrittum, Ahmad Qamar, Zoltan Szabo, Lars Buesing, Maneesh Sahani Abstract: We introduce the Locally Linear Latent Variable Model (LL-LVM), a probabilistic model for non-linear manifold discovery that describes a joint distribution over observations, their manifold coordinates and locally linear maps conditioned on a set of neighbourhood relationships. The model allows straightforward variational optimisation of the posterior distribution on coordinates and locally linear maps from the latent space to the observation space given the data. Thus, the LL-LVM encapsulates the local-geometry preserving intuitions that underlie non-probabilistic methods such as locally linear embedding Talk 3: Ricardo Silver will present Bandits with Unobserved Confounders: A Causal Approach by Elias Bareinboim The Multi-Armed Bandit problem constitutes an archetypal setting for sequential http://ftp.cs.ucla.edu/pub/stat_ser/r460.pdf |

iCalendar | csml_id_246.ics |