When FastText Pays Attention: Efficient Estimation of Word Representations using Constrained Positional Weighting

2021/04/30 by Vít Novotný, Michal Štefánik, Eniafe Festus Ayetiran +2 · 6 citations
Computer Science · #Artificial intelligence #Computer science #Language model #Linguistics #Machine learning #Machine translation #Natural Language Processing Techniques #Natural language processing #Pattern recognition (psychology) #Task (project management) #Text Readability and Simplification #Topic Modeling #Weighting #Word (group theory) #acm:68T50 #cs.CL #msc:68T50

paper · pdf · open access · doi:10.3897/jucs.69619

published in JUCS - Journal of Universal Computer Science 28(2), 181-201 (Verlag der Technischen Universität Graz)

openalex created_date 2021/04/26 · arxiv created 2022/02/28 · openalex publication_date 2022/02/28 · arxiv updated 2022/03/01 · openalex updated_date 2026/08/05

Abstract

In 2018, Mikolov et al. introduced the positional language model, which has characteristics of attention-based neural machine translation models and which achieved state-of-the-art performance on the intrinsic word analogy task. However, the positional model is not practically fast and it has never been evaluated on qualitative criteria or extrinsic tasks. We propose a constrained positional model, which adapts the sparse attention mechanism from neural machine translation to improve the speed of the positional model. We evaluate the positional and constrained positional models on three novel qualitative criteria and on language modeling. We show that the positional and constrained positional models contain interpretable information about the grammatical properties of words and outperform other shallow models on language modeling. We also show that our constrained model outperforms the positional model on language modeling and trains twice as fast.

Citations

• Efficient Estimation of Word Representations in Vector Space
• Sequence to Sequence Learning with Neural Networks
• Numeracy and Decision Making
• Distributed Representations of Words and Phrases and their Compositionality
• Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
• Random texts exhibit Zipf's-law-like word frequency distribution
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• Enriching Word Vectors with Subword Information
• Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling
• The exp-normal distribution is infinitely divisible
• On the Turing Completeness of Modern Neural Network Architectures
• Generating Long Sequences with Sparse Transformers
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• HOGWILD!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
• Are Transformers universal approximators of sequence-to-sequence functions?
• Longformer: The Long-Document Transformer
• Text classification with word embedding regularization and soft similarity measure
• Exploiting Generative AI to Scale up Intelligent Tutoring Systems