Abstract
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.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
- 2.
In this work we refer to recursive neural networks as RecNN to avoid name clash with RNNs.
- 3.
References
Bengio, Y., Ducharme, R., Vincent, P., Jauvin, C.: A neural probabilistic language model. JMLR 3, 1137–1155 (2003)
Coates, A., Ng, A.Y.: Learning feature representations with K-means. In: Montavon, G., Orr, G.B., Müller, K.-R. (eds.) Neural Networks: Tricks of the Trade. LNCS, vol. 7700, pp. 561–580. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-35289-8_30
Collobert, R., Weston, J.: A unified architecture for natural language processing: deep neural networks with multitask learning. In: ICML, pp. 160–167 (2008)
Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., Kuksa, P.: Natural language processing (almost) from scratch. JMLR 12, 2493–2537 (2011)
Conneau, A., Kiela, D., Schwenk, H., Barrault, L., Bordes, A.: Supervised learning of universal sentence representations from natural language inference data. In: EMNLP, pp. 670–680 (2017)
Cormack, G.V., Grossman, M.R., Hedin, B., Oard, D.W.: Overview of the TREC 2010 legal track. In: TREC (2010)
Forgy, E.W.: Cluster analysis of multivariate data: efficiency versus interpretability of classification. Biometrics 21(3), 768–769 (1965)
Freund, Y., Schapire, R.E.: Experiments with a new boosting algorithm. In: ICML, pp. 148–156 (1996)
Goller, C., Kuchler, A.: Learning task-dependent distributed representations by backpropagation through structure. In: IEEE ICNN, pp. 347–352 (1996)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge (2016)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
İrsoy, O., Cardie, C.: Deep recursive neural networks for compositionality in language. In: NIPS, pp. 2096–2104 (2014)
Kim, Y.: Convolutional neural networks for sentence classification. In: EMNLP, pp. 1746–1751 (2014)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: ICLR (2015)
Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. In: ICLR (2017)
Kiss, T., Strunk, J.: Unsupervised multilingual sentence boundary detection. Comput. Linguist. 32(4), 485–525 (2006)
Klein, D., Manning, C.D.: Accurate unlexicalized parsing. In: ACL, pp. 423–430 (2003)
Klimt, B., Yang, Y.: The enron corpus: a new dataset for email classification research. In: Boulicaut, J.-F., Esposito, F., Giannotti, F., Pedreschi, D. (eds.) ECML 2004. LNCS (LNAI), vol. 3201, pp. 217–226. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-30115-8_22
Kotsiantis, S.B.: Bagging and boosting variants for handling classifications problems: a survey. Knowl. Eng. Rev. 29(1), 78–100 (2014)
Le, Q.V., Mikolov, T.: Distributed representations of sentences and documents. In: ICML, pp. 1188–1196 (2014)
Lloyd, S.: Least squares quantization in PCM. IEEE TIT 28(2), 129–137 (1982)
Loshchilov, I., Hutter, F.: Online batch selection for faster training of neural networks. In: ICLR Workshop (2016)
Mikolov, T., Sutskever, I., Chen, K., Corrado, G.S., Dean, J.: Distributed representations of words and phrases and their compositionality. In: NIPS, pp. 3111–3119 (2013)
Neerbek, J., Assent, I., Dolog, P.: Detecting complex sensitive information via phrase structure in recursive neural networks. In: Phung, D., Tseng, V.S., Webb, G.I., Ho, B., Ganji, M., Rashidi, L. (eds.) PAKDD 2018. LNCS (LNAI), vol. 10939, pp. 373–385. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-93040-4_30
Olvera-López, J.A., Carrasco-Ochoa, J.A., Martínez-Trinidad, J.F., Kittler, J.: A review of instance selection methods. AI Rev. 34(2), 133–143 (2010)
Pennington, J., Socher, R., Manning, C.D.: Glove: global vectors for word representation. In: EMNLP, pp. 1532–1543 (2014)
Rush, A.M., Chopra, S., Weston, J.: A neural attention model for abstractive sentence summarization. In: EMNLP (2015)
Settles, B.: Active learning. Synth. Lect. Artif. Intell. Mach. Learn. 6(1), 1–114 (2012)
Socher, R., Huang, E.H., Pennin, J., Manning, C.D., Ng, A.Y.: Dynamic pooling and unfolding recursive autoencoders for paraphrase detection. In: NIPS, pp. 801–809 (2011)
Socher, R., Manning, C.D., Ng, A.Y.: Learning continuous phrase representations and syntactic parsing with recursive neural networks. In: NIPS Deep Learning and Unsupervised Feature Learning Workshop, pp. 1–9 (2010)
Socher, R., et al.: Recursive deep models for semantic compositionality over a sentiment treebank. In: EMNLP, pp. 1631–1642 (2013)
Taylor, A., Marcus, M., Santorini, B.: The penn treebank: an overview. In: Abeillé, A. (ed.) Treebanks. TLTB, pp. 5–22. Springer, Heidelberg (2003). https://doi.org/10.1007/978-94-010-0201-1_1
Tomlinson, S.: Learning task experiments in the TREC 2010 legal track. In: TREC (2010)
Zhao, Z., Liu, T., Li, B., Du, X.: Cluster-driven model for improved word and text embedding. In: ECAI, pp. 99–106 (2016)
Acknowledgments
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 732240 (Synchronicity Project). The authors would like to thank the anonymous reviewers for valuable comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Neerbek, J., Dolog, P., Assent, I. (2019). Selective Training: A Strategy for Fast Backpropagation on Sentence Embeddings. In: Yang, Q., Zhou, ZH., Gong, Z., Zhang, ML., Huang, SJ. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2019. Lecture Notes in Computer Science(), vol 11441. Springer, Cham. https://doi.org/10.1007/978-3-030-16142-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-16142-2_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-16141-5
Online ISBN: 978-3-030-16142-2
eBook Packages: Computer ScienceComputer Science (R0)