K
Kevin Robert Canini
Researcher at Google
Publications - 25
Citations - 1169
Kevin Robert Canini is an academic researcher from Google. The author has contributed to research in topics: Categorization & Hierarchical Dirichlet process. The author has an hindex of 15, co-authored 25 publications receiving 1043 citations. Previous affiliations of Kevin Robert Canini include University of California, Berkeley & PARC.
Papers
More filters
Proceedings Article
Online Inference of Topics with Latent Dirichlet Allocation
TL;DR: Two related RaoBlackwellized online inference algorithms for the latent Dirichlet allocation (LDA) model – incremental Gibbs samplers and particle filters – are introduced and their runtime and performance are compared to that of existing algorithms.
Proceedings ArticleDOI
Finding Credible Information Sources in Social Networks Based on Content and Social Structure
TL;DR: The study indicates that both the topical content of information sources and social network structure affect source credibility, and designs a novel method of automatically identifying and ranking social network users according to their relevance and expertise for a given topic.
Proceedings Article
Deep Lattice Networks and Partial Monotonic Functions
TL;DR: Monotonic deep lattice networks as discussed by the authors are monotonic with respect to a user-specified set of inputs by alternating layers of linear embeddings, ensembles of lattices, and calibrators with appropriate constraints for monotonicity.
Journal ArticleDOI
Monotonic calibrated interpolated look-up tables
Maya R. Gupta,Andrew Cotter,Jan Pfeifer,Konstantin Voevodski,Kevin Robert Canini,Alexander Mangylov,Wojciech Moczydlowski,Alexander Van Esbroeck +7 more
TL;DR: In this paper, the structural risk minimization framework of lattice regression is extended to monotonic functions by adding linear inequality constraints and jointly learning interpretable calibrations of each feature to normalize continuous features and handle categorical or missing data.
Unifying Rational Models of Categorization via the Hierarchical Dirichlet Process - eScholarship
TL;DR: The authors show that existing rational models of categorization are spe- cial cases of a statistical model called the hierarchical Dirichlet process, which can be used to automatically infer a represen- tation of the appropriate complexity for a given category.