Unsupervised Learning of Word-Sequence Representations from Scratch via Convolutional Tensor Decomposition

TL;DR

This paper introduces a novel unsupervised framework combining convolutional tensor decomposition and deconvolution to learn universal word-sequence embeddings from scratch, addressing challenges of context-awareness and efficiency.

Contribution

It proposes a new two-phased ConvDic+DeconvDec framework that effectively learns word-sequence dictionaries and embeddings without pre-training or external data.

Findings

More accurate and efficient than alternating minimization methods.

Embeddings are effective across multiple downstream tasks.

Framework handles varying sentence lengths without additional complexity.

Abstract

Unsupervised text embeddings extraction is crucial for text understanding in machine learning. Word2Vec and its variants have received substantial success in mapping words with similar syntactic or semantic meaning to vectors close to each other. However, extracting context-aware word-sequence embedding remains a challenging task. Training over large corpus is difficult as labels are difficult to get. More importantly, it is challenging for pre-trained models to obtain word-sequence embeddings that are universally good for all downstream tasks or for any new datasets. We propose a two-phased ConvDic+DeconvDec framework to solve the problem by combining a word-sequence dictionary learning model with a word-sequence embedding decode model. We propose a convolutional tensor decomposition mechanism to learn good word-sequence phrase dictionary in the learning phase. It is proved to be more accurate and much more efficient than the popular alternating minimization method. In the decode phase, we introduce a deconvolution framework that is immune to the problem of varying sentence lengths. The word-sequence embeddings we extracted using ConvDic+DeconvDec are universally good for a few downstream tasks we test on. The framework requires neither pre-training nor prior/outside information.