Timezone: »

Moment Distributionally Robust Tree Structured Prediction
Yeshu Li · Danyal Saeed · Xinhua Zhang · Brian Ziebart · Kevin Gimpel

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #509

Structured prediction of tree-shaped objects is heavily studied under the name of syntactic dependency parsing. Current practice based on maximum likelihood or margin is either agnostic to or inconsistent with the evaluation loss. Risk minimization alleviates the discrepancy between training and test objectives but typically induces a non-convex problem. These approaches adopt explicit regularization to combat overfitting without probabilistic interpretation. We propose a moment-based distributionally robust optimization approach for tree structured prediction, where the worst-case expected loss over a set of distributions within bounded moment divergence from the empirical distribution is minimized. We develop efficient algorithms for arborescences and other variants of trees. We derive Fisher consistency, convergence rates and generalization bounds for our proposed method. We evaluate its empirical effectiveness on dependency parsing benchmarks.

Author Information

Yeshu Li (University of Illinois, Chicago)
Danyal Saeed (University of Illinois at Chicago)
Xinhua Zhang (University of Illinois at Chicago (UIC))
Brian Ziebart (University of Illinois at Chicago)
Kevin Gimpel

More from the Same Authors