Stacking With Auxiliary Features
This addresses the limitation of ensembling methods in discriminating among models, offering improvements for tasks like slot filling, entity linking, and object detection, though it appears incremental as it builds on existing stacking techniques.
The paper tackles the problem of effectively discriminating among component models in ensembling by proposing stacking with auxiliary features, which learns to fuse relevant information from multiple systems to improve performance, achieving new state-of-the-art results on Cold Start Slot Filling and Tri-lingual Entity Discovery and Linking tasks and substantial improvements on ImageNet object detection.
Ensembling methods are well known for improving prediction accuracy. However, they are limited in the sense that they cannot discriminate among component models effectively. In this paper, we propose stacking with auxiliary features that learns to fuse relevant information from multiple systems to improve performance. Auxiliary features enable the stacker to rely on systems that not just agree on an output but also the provenance of the output. We demonstrate our approach on three very different and difficult problems -- the Cold Start Slot Filling, the Tri-lingual Entity Discovery and Linking and the ImageNet object detection tasks. We obtain new state-of-the-art results on the first two tasks and substantial improvements on the detection task, thus verifying the power and generality of our approach.