COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring. |
University of Cambridge > Talks.cam > Machine Learning Reading Group @ CUED > Model selection in a large compositional space
Model selection in a large compositional spaceAdd to your list(s) Download to your calendar using vCal
If you have a question about this talk, please contact Konstantina Palla. note: the first 45 mins will be planning and the talk will start at around 3:15pm We often build complex probabilistic models by “composing” simpler models—using one model to generate the latent variables for another model. This allows us to express complex distributions over the observed data and to share statistical structure between different parts of a model. I’ll present a space of matrix decomposition models defined by the composition of a small number of motifs of probabilistic modeling, such as clustering, low rank factorizations, and binary latent factor models. This compositional structure can be represented by a context-free grammar whose production rules correspond to these motifs. By exploiting the structure of this grammar, we can generically and efficiently infer latent components and estimate predictive likelihood for nearly 2500 model structures using a small toolbox of reusable algorithms. Using a greedy search over this grammar, we automatically choose the decomposition structure from raw data by evaluating only a small fraction of all models. The proposed method typically finds the correct structure for synthetic data and backs off gracefully to simpler models under heavy noise. It learns sensible structures for datasets as diverse as image patches, motion capture, 20 Questions, and U.S. Senate votes, all using exactly the same code. I’ll briefly describe my ongoing work on estimating marginal likelihood in this space of models and how I think this work relates to compositional models more generally. This talk is part of the Machine Learning Reading Group @ CUED series. This talk is included in these lists:
Note that ex-directory lists are not shown. |
Other listsArrol Adam Lecture Series Associative experiment Cambridge Genomic Services Seminars SciBar CMS seminars VilamoviceanOther talksMolecular mechanisms of cardiomyopathies in patients with severe non-ischemic heart failure Mechanical properties of cells or cell components on the micro- and nanometer scale Dr Michael Hastings: Circadian Rhythms CANCELLED Jennifer Luff: Secrets, Lies, and the 'Special Relationship' in the Early Cold War Systems for Big Data Applications:Revolutionising personal computing The Warsaw Uprising in Polish Popular Culture after 1989 An SU(3) variant of instanton homology for webs Graph Legendrians and SL2 local systems A polyfold lab report A rose by any other name Cambridge - Corporate Finance Theory Symposium September 2017 - Day 1 St Catharine’s Political Economy Seminar - ‘Global Imbalances and Greece's Exit from the Crisis’ by Dimitrios Tsomocos |