Skip to main content
SpringerLink
Log in

Batch mode active learning via adaptive criteria weights

  • Published:
Applied Intelligence Aims and scope Submit manuscript

A Correction to this article was published on 29 December 2020

This article has been updated

Abstract

Batch mode active learning (BMAL) is absorbed in training reliable classifier with deficient labeled examples by efficiently querying the most valuable unlabeled examples for supervision. In particular, BMAL always selects examples based on the decent-designed criteria, such as (un)certainty and representativeness, etc. However, existing BMAL approaches make a naive trade-off between the criteria and simply combine them with fixed weights, which may yield suboptimal batch selection since the criteria of unlabeled examples would fluctuate after retraining classifier with the newly augmented training set as the learning of classifier progresses. Instead, the weights of the criteria should be assigned properly. To overcome this problem, this paper proposes a novel A daptive C riteria W eights active learning method, abbreviated ACW, which dynamically combines the example selection criteria together to select critical examples for semi-supervised classification. Concretely, we first assign an initial value to each criterion weight, then the current optimal batch is picked from unlabeled pool. Thereafter, the criteria weights are learned and adjusted adaptively by minimizing the objective function with the selected batch at each round. To the best of our knowledge, this work is the first attempt to explore adaptive criteria weights in the context of active learning. The superiority of ACW against the existing state-of-the-art BMAL approaches has also been validated by extensive experimental results on widely used datasets.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Change history

Notes

  1. https://archive.ics.uci.edu/ml/datasets.php

  2. http://www.cad.zju.edu.cn/home/dengcai/Data/MLData.html

References

  1. Ash JT, Zhang C, Krishnamurthy A, Langford J, Agarwal A (2019) Deep batch active learning by diverse, uncertain gradient lower bounds arXiv: Learning

  2. Balcan MF, Broder A, Zhang T (2007) Margin based active learning. In: International conference on computational learning theory. Springer, pp 35–50

  3. Beluch WH, Genewein T, Nürnberger A, Köhler JM (2018) The power of ensembles for active learning in image classification. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 9368–9377

  4. Borgwardt KM, Gretton A, Rasch MJ, Kriegel HP, Schölkopf B, Smola A (2006) Integrating structured biological data by kernel maximum mean discrepancy. Bioinformatics 22(14):e49–e57

    Article  Google Scholar 

  5. Brinker K (2003) Incorporating diversity in active learning with support vector machines. In: Proceedings of the 20th international conference on machine learning (ICML-03), pp 59–66

  6. Cai W, Zhang Y, Zhang Y, Zhou S, Wang W, Chen Z, Ding C (2017) Active learning for classification with maximum model change. ACM Trans Inform Sys (TOIS) 36(2):15

    Google Scholar 

  7. Chakraborty S, Balasubramanian V, Panchanathan S (2015) Adaptive batch mode active learning. IEEE Trans Neural Netw Learning Sys 26(8):1747–1760

    Article  MathSciNet  Google Scholar 

  8. Chakraborty S, Balasubramanian V, Sun Q, Panchanathan S, Ye J (2015) Active batch selection via convex relaxations with guaranteed solution bounds. IEEE Trans Pattern Anal Mach Intell 37 (10):1945–1958

    Article  Google Scholar 

  9. Chattopadhyay R, Fan W, Davidson I, Panchanathan S, Ye J (2013) Joint transfer and batch-mode active learning. In: International conference on machine learning, pp 253–261

  10. Chattopadhyay R, Wang Z, Fan W, Davidson I, Panchanathan S, Ye J (2013) Batch mode active sampling based on marginal probability distribution matching. ACM Transactions on Knowledge Discovery from Data (TKDD) 7(3):13

    Article  Google Scholar 

  11. Deng J, Dong W, Socher R, Li L, Li K, Feifei L (2009) Imagenet: a large-scale hierarchical image database. In: Proceedings of the IEEE conference on computer vision and pattern recognition. IEEE, pp 248–255

  12. Dhurandhar A, Sankaranarayanan K (2015) Improving classification performance through selective instance completion. Mach Learn 100(2-3):425–447

    Article  MathSciNet  Google Scholar 

  13. Donmez P, Carbonell J, Bennett PN (2007) Dual strategy active learning. In: European conference on machine learning. Springer, pp 116–127

  14. Du B, Wang Z, Zhang L, Zhang L, Liu W, Shen J, Tao D (2017) Exploring representativeness and informativeness for active learning. IEEE Trans Cybern 47(1):14–26

    Article  Google Scholar 

  15. Esna Ashari Z, Ghasemzadeh H (2019) Mindful active learning. In: Proceedings of the twenty-eighth international joint conference on artificial intelligence, pp 2265–2271

  16. Everingham M, Van Gool L, Williams CKI, Winn J, Zisserman A (2010) The pascal visual object classes (voc) challenge. Int J Comput Vis 88(2):303–338

    Article  Google Scholar 

  17. Freund Y, Seung HS, Shamir E, Tishby N (1997) Selective sampling using the query by committee algorithm. Machine Learning 28(2-3):133–168

    Article  Google Scholar 

  18. Fu W, Wang M, Hao S, Wu X (2018) Scalable active learning by approximated error reduction. In: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining. ACM, pp 1396–1405

  19. Fujita H, Gaeta A, Loia V, Orciuoli F (2019) Improving awareness in early stages of security analysis: a zone partition method based on grc. Appl Intell 49:1063–1077

    Article  Google Scholar 

  20. Fujita H, Gaeta A, Loia V, Orciuoli F (2019) Resilience analysis of critical infrastructures: a cognitive approach based on granular computing. IEEE Trans Sys Man Cybern 49(5):1835–1848

    Google Scholar 

  21. Gilad-Bachrach R, Navot A, Tishby N (2006) Query by committee made real. In: Advances in neural information processing systems, pp 443–450

  22. Gretton A, Borgwardt KM, Rasch MJ, Schölkopf B, Smola A (2012) A kernel two-sample test. J Mach Learn Res 13(Mar):723–773

    MathSciNet  MATH  Google Scholar 

  23. Guo Y, Greiner R (2007) Optimistic active-learning using mutual information. In: IJCAI, vol 7, pp 823–829

  24. Guo Y, Schuurmans D (2008) Discriminative batch mode active learning. In: Advances in neural information processing systems, pp 593–600

  25. He X (2009) Laplacian regularized d-optimal design for active learning and its application to image retrieval. IEEE Trans Image Process 19(1):254–263

    MathSciNet  MATH  Google Scholar 

  26. Hoi SC, Jin R, Lyu MR (2009) Batch mode active learning with applications to text categorization and image retrieval. IEEE Trans Knowl Data Eng 21(9):1233–1248

    Article  Google Scholar 

  27. Hoi SC, Jin R, Zhu J, Lyu MR (2009) Semisupervised svm batch mode active learning with applications to image retrieval. ACM Transactions on Information Systems (TOIS) 27(3):16

    Article  Google Scholar 

  28. Hu Y, Zhang D, Jin Z, Cai D, He X (2013) Active learning via neighborhood reconstruction. In: Proceedings of the twenty-third international joint conference on artificial intelligence, pp 1415–1421

  29. Huang SJ, Jin R, Zhou ZH (2014) Active learning by querying informative and representative examples. IEEE Trans Pattern Anal Mach Intell 36(10):1936–1949

    Article  Google Scholar 

  30. Joshi AJ, Porikli F, Papanikolopoulos N (2009) Multi-class active learning for image classification. In: 2009 IEEE conference on computer vision and pattern recognition. IEEE, pp 2372–2379

  31. Kapoor A, Horvitz E, Basu S (2007) Selective supervision: guiding supervised learning with decision-theoretic active learning. In: IJCAI, vol 7, pp 877–882

  32. Kirsch A, van Amersfoort J, Gal Y (2019) Batchbald: efficient and diverse batch acquisition for deep bayesian active learning. In: Advances in neural information processing systems, pp 7026–7037

  33. Konyushkova K, Sznitman R, Fua P (2017) Learning active learning from data. In: Advances in neural information processing systems 30, pp 4225–4235

  34. Li CL, Ferng CS, Lin HT (2012) Active learning with hinted support vector machine. In: Proceedings of asian conference on machine learning. PMLR, pp 221–235

  35. Li Y, Wang Y, Yu DJ, Ye N, Hu P, Zhao R (2019) Ascent: active supervision for semi-supervised learning. IEEE Trans Knowl Data Eng 32(5):868–882

    Article  Google Scholar 

  36. Long C, Hua G (2015) Multi-class multi-annotator active learning with robust gaussian process for visual recognition. In: Proceedings of the IEEE international conference on computer vision, pp 2839–2847

  37. Long C, Hua G, Kapoor A (2016) A joint gaussian process model for active visual recognition with expertise estimation in crowdsourcing. Int J Comput Vis 116(2):136–160

    Article  MathSciNet  Google Scholar 

  38. Mallapragada PK, Jin R, Jain AK (2008) Active query selection for semi-supervised clustering. In: 2008 19Th international conference on pattern recognition, pp 1–4. IEEE

  39. Murugesan K, Carbonell J (2017) Active learning from peers. In: Advances in neural information processing systems 30, pp 7008–7017

  40. Nie F, Wang H, Huang H, Ding C (2013) Early active learning via robust representation and structured sparsity. In: Twenty-third international joint conference on artificial intelligence, pp 1572–1578

  41. Nie F, Wang X, Huang H (2014) Clustering and projected clustering with adaptive neighbors. In: Proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining, pp 977–986

  42. Paul S, Bappy JH, Roy-Chowdhury AK (2017) Non-uniform subset selection for active learning in structured data. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 6846–6855

  43. Pinsler R, Gordon J, Nalisnick E, Hernandezlobato JM (2019) Bayesian batch active learning as sparse subset approximation. In: Advances in neural information processing systems, pp 6356–6367

  44. Sandoval AM, Diaz J, Llanos LC, Redondo T (2019) Biomedical term extraction: Nlp techniques in computational medicine. Int J Interact Multimed Artificial Intell 5(4):51–59

    Google Scholar 

  45. Settles B (2009) Active learning literature survey. Tech. rep., University of Wisconsin-Madison Department of Computer Sciences

  46. Seung HS, Opper M, Sompolinsky H (1992) Query by committee. In: Proceedings of the fifth annual workshop on computational learning theory, pp 287–294. ACM

  47. Tang YP, Huang SJ (2019) Self-paced active learning: query the right thing at the right time. In: Proceedings of the AAAI conference on artificial intelligence, vol 33, pp 5117– 5124

  48. Tang YP, Li GX, Huang SJ (2019) ALiPy: active learning in python. Tech. rep., Nanjing University of Aeronautics and Astronautics. https://github.com/NUAA-AL/ALiPy. Available as arXiv:1901.03802

  49. Tong S, Koller D (2001) Support vector machine active learning with applications to text classification. J Mach Learn Res 2(Nov):45–66

    MATH  Google Scholar 

  50. Vasisht D, Damianou A, Varma M, Kapoor A (2014) Active learning for sparse bayesian multilabel classification. In: Proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining. ACM, pp 472–481

  51. Wang H, Chang X, Shi L, Yang Y, Shen YD (2018) Uncertainty sampling for action recognition via maximizing expected average precision. In: IJCAI, pp 964–970

  52. Wang H, Du L, Zhou P, Shi L, Shen YD (2015) Convex batch mode active sampling via α-relative pearson divergence. In: Twenty-ninth AAAI conference on artificial intelligence

  53. Wang H, Zhou R, Shen YD (2019) Bounding uncertainty for active batch selection. In: Proceedings of the AAAI conference on artificial intelligence, vol 33, pp 5240–5247

  54. Wang K, Zhang D, Li Y, Zhang R, Lin L (2017) Cost-effective active learning for deep image classification. IEEE Trans Circ Sys Video Technol 27(12):2591–2600

    Article  Google Scholar 

  55. Wang Z, Ye J (2015) Querying discriminative and representative samples for batch mode active learning. ACM Trans Knowl Discovery Data (TKDD) 9(3):17

    Google Scholar 

  56. Wu Y, Wu Q, Dey N, Sherratt S (2020) Learning models for semantic classification of insufficient plantar pressure images. Int J Interact Multimed Artificial Intell 6(1):51–61

    Google Scholar 

  57. Xiong C, Johnson DM, Corso JJ (2017) Active clustering with model-based uncertainty reduction. IEEE Trans Pattern Anal Machine Intell 39(1):5–17

    Article  Google Scholar 

  58. Xu Z, Yu K, Tresp V, Xu X, Wang J (2003) Representative sampling for text classification using support vector machines. In: European conference on information retrieval. Springer, pp 393–407

  59. Yang Y, Ma Z, Nie F, Chang X, Hauptmann AG (2015) Multi-class active learning by uncertainty sampling with diversity maximization. Int J Comput Vis 113(2):113–127

    Article  MathSciNet  Google Scholar 

  60. Yin C, Qian B, Cao S, Li X, Wei J, Zheng Q, Davidson I (2017) Deep similarity-based batch mode active learning with exploration-exploitation. In: 2017 IEEE international conference on data mining (ICDM). IEEE, pp 575–584

  61. Yoo D, Kweon IS (2019) Learning loss for active learning. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 93–102

  62. Zhang Y, Lease M, Wallace BC (2017) Active discriminative text representation learning. In: Thirty-first AAAI conference on artificial intelligence

  63. Zuluaga M, Sergent G, Krause A, Püschel M (2013) Active learning for multi-objective optimization. In: International conference on machine learning, pp 462–470

Download references

Acknowledgements

The authors would like to thank the anonymous reviewers for their valuable comments and suggestions. This article has been awarded by the National Natural Science Foundation of China (61941113), the Fundamental Research Fund for the Central Universities (30918015103, 30918012204), Nanjing Science and Technology Development Plan Project (201805036), and “13th Five-Year” equipment field fund (61403120501), China Academy of Engineering Consulting Research Project (2019-ZD-1-02-02), National Social Science Foundation (18BTQ073), State Grid Technology Project (5211XT190033).

Author information

Authors and Affiliations

  1. School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China

    Hao Li, Yongli Wang, Peng Hu & Ruxin Zhao

  2. School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China

    Yanchao Li

  3. Science and technology on Complex Systems Simulation Laboratory, Nanjing, China

    Gang Xiao

Authors
  1. Hao Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yongli Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yanchao Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Gang Xiao
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Peng Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Ruxin Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yongli Wang.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The original online version of this article was revised: The affiliation of the second and third authors were incorrect.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, H., Wang, Y., Li, Y. et al. Batch mode active learning via adaptive criteria weights. Appl Intell 51, 3475–3489 (2021). https://doi.org/10.1007/s10489-020-01953-4

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-020-01953-4

Keywords

Search

Navigation