Automated Concatenation of Embeddings for Structured Prediction
Abstract: Pretrained contextualized embeddings are powerful word representations for structured prediction tasks. Recent work found that better word representations can be obtained by concatenating different types of embeddings. However, the selection of embeddings to form the best concatenated representation usually varies depending on the task and the collection of candidate embeddings, and the ever-increasing number of embedding types makes it a more difficult problem. In this paper, we propose Automated Concatenation of Embeddings (ACE) to automate the process of finding better concatenations of embeddings for structured prediction tasks, based on a formulation inspired by recent progress on neural architecture search. Specifically, a controller alternately samples a concatenation of embeddings, according to its current belief of the effectiveness of individual embedding types in consideration for a task, and updates the belief based on a reward. We follow strategies in reinforcement learning to optimize the parameters of the controller and compute the reward based on the accuracy of a task model, which is fed with the sampled concatenation as input and trained on a task dataset. Empirical results on 6 tasks and 21 datasets show that our approach outperforms strong baselines and achieves state-of-the-art performance with fine-tuned embeddings in the vast majority of evaluations.
One-sentence Summary: We propose ACE, which automatically searches the best word embedding concatenation as word representation. ACE achieved state-of-the-art results on 6 structured prediction tasks over 19 out of 21 datasets.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Community Implementations: [ 4 code implementations](https://www.catalyzex.com/paper/arxiv:2010.05006/code)
Reviewed Version (pdf): https://openreview.net/references/pdf?id=PZnmQnazBN
13 Replies
Loading