The LCFI website uses cookies only for anonymised website statistics and for ensuring our security, never for tracking or identifying you individually. To find out more, and to find out how we protect your personal information, please read our privacy policy.


Conference Paper by Mark Rowland, Adrian Weller

Uprooting and Rerooting Higher-Order Graphical Models

31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA

The idea of uprooting and rerooting graphical models was introduced specifically for binary pairwise models by Weller as a way to transform a model to any of a whole equivalence class of related models, such that inference on any one model yields inference results for all others. This is very helpful since inference, or relevant bounds, may be much easier to obtain or more accurate for some model in the class. Here we introduce methods to extend the approach to models with higher-order potentials and develop theoretical insights. In particular, we show that the triplet-consistent polytope TRI is unique in being ‘universally rooted’. We demonstrate empirically that rerooting can significantly improve accuracy of methods of inference for higher-order models at negligible computational cost.

Download Conference Paper