Unsupervised Learning of Word-Sequence Representations from Scratch via Convolutional Tensor Decomposition
This addresses the problem of generating versatile text embeddings for NLP applications, though it appears incremental as it builds on existing unsupervised methods.
The paper tackles the challenge of extracting context-aware word-sequence embeddings without supervision by proposing a two-phased framework combining convolutional tensor decomposition for dictionary learning and a deconvolution decode model, resulting in embeddings that are universally good for tested downstream tasks without requiring pre-training or external data.
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.