Learning to Speed Up Structured Output Prediction

TL;DR

This paper introduces a method to accelerate structured output prediction by training a speedup classifier that mimics a black-box model, significantly reducing prediction time without sacrificing accuracy.

Contribution

It proposes a learning-to-search approach to create a speedup classifier that guides inference efficiently while maintaining the original model's accuracy.

Findings

Speedup classifier outperforms greedy search in speed

No accuracy loss observed with the speedup classifier

Effective in entity and relation extraction tasks

Abstract

Predicting structured outputs can be computationally onerous due to the combinatorially large output spaces. In this paper, we focus on reducing the prediction time of a trained black-box structured classifier without losing accuracy. To do so, we train a speedup classifier that learns to mimic a black-box classifier under the learning-to-search approach. As the structured classifier predicts more examples, the speedup classifier will operate as a learned heuristic to guide search to favorable regions of the output space. We present a mistake bound for the speedup classifier and identify inference situations where it can independently make correct judgments without input features. We evaluate our method on the task of entity and relation extraction and show that the speedup classifier outperforms even greedy search in terms of speed without loss of accuracy.