A Mutual Information Maximization Perspective of Language Representation Learning
It provides a theoretical unification for language representation learning, potentially enabling knowledge transfer across domains like NLP and computer vision.
The paper shows that state-of-the-art word representation learning methods maximize a lower bound on mutual information between parts of sentences, unifying classical and modern embeddings, and introduces a new self-supervised objective based on this framework.
We show state-of-the-art word representation learning methods maximize an objective function that is a lower bound on the mutual information between different parts of a word sequence (i.e., a sentence). Our formulation provides an alternative perspective that unifies classical word embedding models (e.g., Skip-gram) and modern contextual embeddings (e.g., BERT, XLNet). In addition to enhancing our theoretical understanding of these methods, our derivation leads to a principled framework that can be used to construct new self-supervised tasks. We provide an example by drawing inspirations from related methods based on mutual information maximization that have been successful in computer vision, and introduce a simple self-supervised objective that maximizes the mutual information between a global sentence representation and n-grams in the sentence. Our analysis offers a holistic view of representation learning methods to transfer knowledge and translate progress across multiple domains (e.g., natural language processing, computer vision, audio processing).