Train and Test Tightness of LP Relaxations in Structured Prediction

@article{Meshi2016TrainAT,
  title={Train and Test Tightness of LP Relaxations in Structured Prediction},
  author={Ofer Meshi and M. Mahdavi and Adrian Weller and D. Sontag},
  journal={J. Mach. Learn. Res.},
  year={2016},
  volume={20},
  pages={13:1-13:34}
}
Structured prediction is used in areas such as computer vision and natural language processing to predict structured outputs such as segmentations or parse trees. In these settings, prediction is performed by MAP inference or, equivalently, by solving an integer linear program. Because of the complex scoring functions required to obtain accurate predictions, both learning and inference typically require the use of approximate solvers. We propose a theoretical explanation to the striking… Expand
Deep Structured Prediction with Nonlinear Output Transformations
Graph Structured Prediction Energy Networks
Exact MAP Inference by Avoiding Fractional Vertices
Argument Mining with Structured SVMs and RNNs
Block Stability for MAP Inference
Learning latent variable structured prediction models with Gaussian perturbations
...
1
2
...

References

SHOWING 1-10 OF 64 REFERENCES
How Hard is Inference for Structured Prediction?
Structured Learning with Approximate Inference
Advanced Structured Prediction
A Linear Programming Formulation for Global Inference in Natural Language Tasks
Learning Max-Margin Tree Predictors
Concise Integer Linear Programming Formulations for Dependency Parsing
Structured Prediction Cascades
...
1
2
3
4
5
...