Data Noising as Smoothing in Neural Network Language Models
This addresses a bottleneck in regularization for NLP tasks, offering a novel method that is incremental but domain-specific.
The paper tackled the lack of effective noising primitives for discrete sequence-level tasks like language modeling by connecting input noising to smoothing in n-gram models, resulting in performance gains in language modeling and machine translation.
Data noising is an effective technique for regularizing neural network models. While noising is widely adopted in application domains such as vision and speech, commonly used noising primitives have not been developed for discrete sequence-level settings such as language modeling. In this paper, we derive a connection between input noising in neural network language models and smoothing in $n$-gram models. Using this connection, we draw upon ideas from smoothing to develop effective noising schemes. We demonstrate performance gains when applying the proposed schemes to language modeling and machine translation. Finally, we provide empirical analysis validating the relationship between noising and smoothing.