# Delta Embedding Learning

## TL;DR

Delta Embedding Learning is a novel method that incrementally fine-tunes word embeddings with regularization, enhancing their semantic quality and improving NLP task performance.

## Contribution

It introduces a structured regularization technique for incremental embedding tuning, addressing limitations of unsupervised embeddings in NLP.

## Key findings

- Consistent performance improvements across multiple NLP tasks
- Enhanced semantic properties of tuned embeddings
- Effective incremental tuning without forgetting previous information

## Abstract

Unsupervised word embeddings have become a popular approach of word representation in NLP tasks. However there are limitations to the semantics represented by unsupervised embeddings, and inadequate fine-tuning of embeddings can lead to suboptimal performance. We propose a novel learning technique called Delta Embedding Learning, which can be applied to general NLP tasks to improve performance by optimized tuning of the word embeddings. A structured regularization is applied to the embeddings to ensure they are tuned in an incremental way. As a result, the tuned word embeddings become better word representations by absorbing semantic information from supervision without "forgetting." We apply the method to various NLP tasks and see a consistent improvement in performance. Evaluation also confirms the tuned word embeddings have better semantic properties.