Poster
Deep Structured Prediction with Nonlinear Output Transformations
Colin Graber · Ofer Meshi · Alexander Schwing

Thu Dec 6th 10:45 AM -- 12:45 PM @ Room 517 AB #139

Deep structured models are widely used for tasks like semantic segmentation, where explicit correlations between variables provide important prior information which generally helps to reduce the data needs of deep nets. However, current deep structured models are restricted by oftentimes very local neighborhood structure, which cannot be increased for computational complexity reasons, and by the fact that the output configuration, or a representation thereof, cannot be transformed further. Very recent approaches which address those issues include graphical model inference inside deep nets so as to permit subsequent non-linear output space transformations. However, optimization of those formulations is challenging and not well understood. Here, we develop a novel model which generalizes existing approaches, such as structured prediction energy networks, and discuss a formulation which maintains applicability of existing inference techniques.

Author Information

Colin Graber (University of Illinois at Urbana-Champaign)
Ofer Meshi (Google)
Alex Schwing (University of Illinois at Urbana-Champaign)

More from the Same Authors