Train and Test Tightness of LP Relaxations in Structured Prediction.

Ofer Meshi Mehrdad Mahdavi Adrian Weller David A. Sontag

Published in: ICML (2016)

Keyphrases

structured prediction
maximum margin
conditional random fields
lp relaxation
markov networks
linear programming
latent variables
message passing
convex optimization
efficient learning
learning algorithm
belief networks
parameter estimation
information extraction
hidden markov models
pairwise
lower bound