|COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring.|
An Introduction to Sum Product Networks
If you have a question about this talk, please contact Colorado Reed.
Sum product networks (SPNs) are a new family of deep probabilistic models in which exact inference is tractable. SPNs are directed acyclic graphs with variables as leaves, sums and products as internal nodes, and weighted edges. A SPN is an arithmetic circuit which under some conditions (completeness and consistency) represents the partition function and all marginals of some graphical model. Essentially all tractable graphical models can be cast as SPNs, but SPNs are also more general. Discriminative and generative learning of SPNs can be efficiently implemented using hard EM and hard gradient descent. These methods avoid the problem of gradient diffusion in deep architectures and allow us to effectively work with SPNs of more than 30 layers of hidden variables. Several experiments show that SPNs have state of the art performance on different image completion and classification tasks, outperforming alternative deep and shallow methods.
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 listsCambridge Carbon Footprint Meeting the Challenge of Healthy Ageing in the 21st Century Geometry in Science
Other talksLies, Damn'd lies and statistics: why it is (almost) impossible to communicate risk ethically Mapping regulatory variation in human cells TBC Challenging bariatric care practices and fat stigma in ICU Personal data: Innovation and Analytics Exploring psychotic experiences in ‘non-need for care’ populations: Findings from the UNIQUE study