Efficient Multi-Template Learning for Structured Prediction

TL;DR

This paper introduces a novel multi-template learning approach for structured prediction that automatically learns the importance of each template, improving performance and efficiency over existing methods like CRFs and Structural SVMs.

Contribution

The paper proposes a new multiple template learning paradigm formulated as a multiple kernel learning problem, with an efficient primal cutting plane algorithm, applicable to structured prediction tasks.

Findings

Outperforms CRFs and Structural SVMs in experiments

More efficient than OnlineMKL on sparse, high-dimensional data

Extends to generalized p-block norm regularization with competitive results

Abstract

Conditional random field (CRF) and Structural Support Vector Machine (Structural SVM) are two state-of-the-art methods for structured prediction which captures the interdependencies among output variables. The success of these methods is attributed to the fact that their discriminative models are able to account for overlapping features on the whole input observations. These features are usually generated by applying a given set of templates on labeled data, but improper templates may lead to degraded performance. To alleviate this issue, in this paper, we propose a novel multiple template learning paradigm to learn structured prediction and the importance of each template simultaneously, so that hundreds of arbitrary templates could be added into the learning model without caution. This paradigm can be formulated as a special multiple kernel learning problem with exponential number of constraints. Then we introduce an efficient cutting plane algorithm to solve this problem in the primal, and its convergence is presented. We also evaluate the proposed learning paradigm on two widely-studied structured prediction tasks, \emph{i.e.} sequence labeling and dependency parsing. Extensive experimental results show that the proposed method outperforms CRFs and Structural SVMs due to exploiting the importance of each template. Our complexity analysis and empirical results also show that our proposed method is more efficient than OnlineMKL on very sparse and high-dimensional data. We further extend this paradigm for structured prediction using generalized $p$-block norm regularization with $p>1$, and experiments show competitive performances when $p \in [1,2)$.