A comparative study on regularization strategies for embedding-based neural networks

Abstract

This paper aims to compare different regularization strategies to address a common phenomenon, severe overhtting, in embedding-based neural networks for NLP. We chose two widely studied neural models and tasks as our testbed. We tried several frequently applied or newly proposed regularization strategies, including penalizing weights (embeddings excluded), penalizing embeddings, reembedding words, and dropout. We also emphasized on incremental hyperparameter tuning, and combining different regularizations. The results provide a picture on tuning hyperparameters for neural NLP models.