Selective Training: A Strategy for Fast Backpropagation on Sentence Embeddings

Jan Neerbek, Peter Dolog, Ira Assent

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

1 Citation (Scopus)
39 Downloads (Pure)


Representation or embedding based machine learning models, such as language models or convolutional neural networks have shown great potential for improved performance. However, for complex models on large datasets training time can be extensive, approaching weeks, which is often infeasible in practice. In this work, we present a method to reduce training time substantially by selecting training instances that provide relevant information for training. Selection is based on the similarity of the learned representations over input instances, thus allowing for learning a non-trivial weighting scheme from multi-dimensional representations. We demonstrate the efficiency and effectivity of our approach in several text classification tasks using recursive neural networks. Our experiments show that by removing approximately one fifth of the training data the objective function converges up to six times faster without sacrificing accuracy.

Original languageEnglish
Title of host publicationAdvances in Knowledge Discovery and Data Mining - 23rd Pacific-Asia Conference, PAKDD 2019, Macau, China, April 14-17, 2019, Proceedings, Part III
EditorsQiang Yang, Min-Ling Zhang, Zhiguo Gong, Sheng-Jun Huang, Zhi-Hua Zhou
Number of pages14
PublisherSpringer VS
Publication date2019
ISBN (Print)978-3-030-16141-5
Publication statusPublished - 2019
EventPacific-Asia Conference on Knowledge Discovery and Data Mining - Macau, China
Duration: 14 Apr 201917 Apr 2019
Conference number: 23rd


ConferencePacific-Asia Conference on Knowledge Discovery and Data Mining
SeriesLecture Notes in Computer Science


  • Machine learning
  • Neural network
  • Recursive models
  • Selective training


Dive into the research topics of 'Selective Training: A Strategy for Fast Backpropagation on Sentence Embeddings'. Together they form a unique fingerprint.

Cite this