Timezone: »

Ensembling Graph Predictions for AMR Parsing
Thanh Lam Hoang · Gabriele Picco · Yufang Hou · Young-Suk Lee · Lam Nguyen · Dzung Phan · Vanessa Lopez · Ramon Fernandez Astudillo

Thu Dec 09 08:30 AM -- 10:00 AM (PST) @

In many machine learning tasks, models are trained to predict structure data such as graphs. For example, in natural language processing, it is very common to parse texts into dependency trees or abstract meaning representation (AMR) graphs. On the other hand, ensemble methods combine predictions from multiple models to create a new one that is more robust and accurate than individual predictions. In the literature, there are many ensembling techniques proposed for classification or regression problems, however, ensemble graph prediction has not been studied thoroughly. In this work, we formalize this problem as mining the largest graph that is the most supported by a collection of graph predictions. As the problem is NP-Hard, we propose an efficient heuristic algorithm to approximate the optimal solution. To validate our approach, we carried out experiments in AMR parsing problems. The experimental results demonstrate that the proposed approach can combine the strength of state-of-the-art AMR parsers to create new predictions that are more accurate than any individual models in five standard benchmark datasets.

Author Information

Thanh Lam Hoang (IBM Research)
Gabriele Picco (IBM)
Yufang Hou (Institute for Computational Linguistics, Heidelberg University, Heidelberg University)
Young-Suk Lee (IBM, International Business Machines)
Lam Nguyen (IBM Research, Thomas J. Watson Research Center)
Dzung Phan (IBM Research, T. J. Watson Research Center)
Vanessa Lopez (IBM Research Europe)
Ramon Fernandez Astudillo (IBM Research)

More from the Same Authors