Skip to main content
SpringerLink
Log in

Structured feature selection and task relationship inference for multi-task learning

  • Regular Paper
  • Published:
Knowledge and Information Systems Aims and scope Submit manuscript

Abstract

Multi-task learning (MTL) aims to enhance the generalization performance of supervised regression or classification by learning multiple related tasks simultaneously. In this paper, we aim to extend the current MTL techniques to high dimensional data sets with structured input and structured output (SISO), where the SI means the input features are structured and the SO means the tasks are structured. We investigate a completely ignored problem in MTL with SISO data: the interplay of structured feature selection and task relationship modeling. We hypothesize that combining the structure information of features and task relationship inference enables us to build more accurate MTL models. Based on the hypothesis, we have designed an efficient learning algorithm, in which we utilize a task covariance matrix related to the model parameters to capture the task relationship. In addition, we design a regularization formulation for incorporating the structured input features in MTL. We have developed an efficient iterative optimization algorithm to solve the corresponding optimization problem. Our algorithm is based on the accelerated first order gradient method in conjunction with the projected gradient scheme. Using two real-world data sets, we demonstrate the utility of the proposed learning methods.

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

Similar content being viewed by others

Notes

  1. GBM, glioblastoma; AO, anaplastic oligodendroglioma.

  2. Stability score measures the variation in voxel activity across the 58 training stimuli.

  3. MTLPTR selects the same feature set for all tasks due to the block regularization, hence a single number is reported.

  4. The 13 genes’ KEGG IDs: 534, 535, 536, 1067, 1068, 2211, 2224, 2229, 2284, 3262, 4978, 5068, 7190.

References

  1. Argyriou A, Micchelli A, Pontil M, Ying Y (2007) A spectral regularization framework for multi-task structure learning. In: NIPS

  2. Argyriou A, Evgeniou T, Pontil M (2006) Multi-task feature learning. In: NIPS

  3. Bakker B, Heskes T (2003) Task clustering and gating for bayesian multitask learning. J Mach Learn Res 4:83–99

    Google Scholar 

  4. Bolón-Canedo V, Snchez-Maroón N, Alonso-Betanzos A (2012) A review of feature selection methods on synthetic data. Knowledge and information systems, pp 1–37, ISSN 0219–1377

  5. Bonilla EV, Chai KM, Williams CKI (2007) Multi-task gaussian process prediction. In: NIPS

  6. Bouguila N, Ziou D (2011) A countably infinite mixture model for clustering and feature selection. Knowl Inf Syst, pp 1–20, ISSN 0219–1377. doi:10.1007/s10115-011-0467-4

  7. Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press, Cambridge

    MATH  Google Scholar 

  8. Chang C, Lin C (2001) Libsvm: a library for support vector machines. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm

  9. Chapelle O, Shivaswamy PK, Vadrevu S, Weinberger KQ, Zhang Y, Tseng BL (2010) Multi-task learning for boosting with application to web search ranking. In: KDD, pp 1189–1198

  10. Chen J, Liu J, Jieping Y (2010) Learning incoherent sparse and low-rank patterns from multiple tasks. In: The sixteenth ACM SIGKDD international conference on knowledge discovery and data mining (SIGKDD, 2010)

  11. Chen J, Zhou J, Ye J (2011) Integrating low-rank and group-sparse structures for robust multi-task learning. In: KDD, pp 42–50

  12. Chen W, Yuan Y, Zhang L (2010) Scalable influence maximization in social networks under the linear threshold model. In: ICDM, pp 88–97

  13. Chen X, Pan W, Kwok JT, Carbonell JG (2009) Accelerated gradient method for multi-task sparse learning problem. In: IEEE international conference on data mining (ICDM09), pp 746–751, ISSN 1550–4786

  14. Evgeniou T, Pontil M (2004) Regularized multi-task learning. In: KDD, pp 109–117

  15. Evgeniou T, Micchelli CA, Pontil M (2005) Learning multiple tasks with kernel methods. J Mach Learn Res 6:615–637

    Google Scholar 

  16. Fei H, Huan J (2010) Boosting with structure information in the functional space: an application to graph classification. In: Proceedings of the ACM SIGKDD international conference on knowledge discovery and data mining (SIGKDD)

  17. Fei H, Quanz B, Huan J (2010) Regularization and feature selection for networked features. In: Proceedings of the 19th ACM conference on information and, knowledge management (CIKM’10)

  18. Fellbaum C (1998) WordNet: an electronic lexical database. The MIT Press, Cambridge

    MATH  Google Scholar 

  19. Gomez J-C, Boiy E, Moens M-F (2012) Highly discriminative statistical features for email classification. Knowl Inf Syst 31(1):23–53

    Article  Google Scholar 

  20. Jacob L, Bach F, Vert J-P (2009) Clustered multi-task learning: a convex formulation. In: Advances in neural information processing systems (NIPS09)

  21. Jalali A, Ravikumar P, Sanghavi S, Ruan C (2010) A dirty model for multi-task learning. In: Proceedings of the advances in neural information processing systems (NIPS 2010), pp 964–972

  22. Ji S, Ye J (2009) An accelerated gradient method for trace norm minimization. In: ICML ’09: Proceedings of the 26th annual international conference on machine learning. ACM, New York, NY, USA, pp 457–464, ISBN 978-1-60558-516-1. http://doi.acm.org/10.1145/1553374.1553434

  23. Kanehisa M, Goto S, Hattori M, Aoki-Kinoshita KF, Itoh M, Kawashima S, Katayama T, Araki M, Hirakawa M (2006) From genomics to chemical genomics: new developments in KEGG. Nucleic Acids Res 34:D354–D357

    Google Scholar 

  24. Kato T, Kashima H, Asai K, Sugiyama M (2007) Multi-task learning via conic programming. In: NIPS

  25. Kong X, Yu PS (2012) gMLC: a multi-label feature selection framework for graph classification. Knowl Inf Syst 31(2):281–305

    Article  Google Scholar 

  26. Levy S (2001) Interactive 3-d visualization of particle systems with partiview. In: Proceedings of the international astronomical union symposium on astrophysical supercomputing using particles, vol 208

  27. Li C, Li H (2008) Network-constrained regularization and variable selection for analysis of genomic data. Bioinformatics 24(9):1175–1182

    Article  Google Scholar 

  28. Liu H, Palatucci M, Zhang J (2009) Blockwise coordinate descent procedures for the multi-task lasso, with applications to neural semantic basis discovery. In: Proceedings of the 26th international conference on machine learning (ICML 2009)

  29. Liu J, Ji S, Ye J (2009) Multi-task feature learning via efficient \(\ell _{2,1}\)-norm minimization. In: Uncertainty in artificial intelligence

  30. Lounici K, Pontil M, Van De Geer S (2009) Taking advantage of sparsity in multi-task learning. Knowl Inf Syst 20(1):109–348

    Google Scholar 

  31. Mitchell TM, Shinkareva SV, Andrew A et al (2008) Predicting human brain activity associated with the meanings of nouns. Science 320:1191–1195

    Article  Google Scholar 

  32. Nesterov Y (2003) Introductory lectures on convex optimization: a basic course, pp 25–32

  33. Nesterov Y (2007) Gradient methods for minimizing composite objective function. CORE Disc Paper 76:265–286

    Google Scholar 

  34. Nutt CL, Mani DR, Betensky RA et al (2003) Gene expression-based classification of malignant gliomas correlates better with survival than histological classification. Cancer Res 63:1602–1607

    Google Scholar 

  35. National Institute of Health (2010) Metastatic cancer. http://www.cancer.gov/cancertopics/factsheet/Sites-Types/metastatic

  36. Quattoni A, Carreras X, Collins M, Darrell T (2009) An efficient projection for l1, infinity regularization. In: Proceedings of the 26th international conference on machine learning (ICML 2009)

  37. Romero DM, Meeder B, Kleinberg JM (2011) Differences in the mechanics of information diffusion across topics: idioms, political hashtags, and complex contagion on twitter. In: WWW, pp 695–704

  38. Sandler T, Talukdar PP, Ungar LH (2008) Regularized learning with networks of features. In: NIPS08

  39. Schwaighofer A, Volker T, Yu K (2004) Learning gaussian process kernels via hierarchical bayes. In: In advances in neural information processing systems, NIPS. MIT Press, pp 1209–1216

  40. Singh D, Febbo PG et al (2002) Gene expression correlates of clinical prostate cancer behavior. Cancer cell 1(2):203–209, ISSN 1535–6108

    Google Scholar 

  41. Su AI, Welsh JB, Sapinoso LM et al (2001) Molecular classification of human carcinomas by use of gene expression signatures. Cancer Res 15(20):7388–7393

    Google Scholar 

  42. Vapnik VN (1995) The nature of statistical learning theory. Springer, Berlin

    Book  MATH  Google Scholar 

  43. Vivanco I, Sawyers CL (2002) The phosphatidylinositol 3-kinase akt pathway in human cancer. Nat Rev Cancer 2(7):489–501

    Article  Google Scholar 

  44. Xu Z, Kersting K (2011) Multi-task learning with task relations. In: ICDM, pp 884–893

  45. Yang J, Leskovec J (2010) Modeling information diffusion in implicit networks. In: ICDM, pp 599–608

  46. Yu H, Yang J (2001) A direct LDA algorithm for high-dimensional data with application to face recognition. Pattern Recognit 34(10):2067–2070

    Article  MATH  Google Scholar 

  47. Yu S, Tresp V, Yu K (2007) Robust multi-task learning with t-processes. In: Proceedings of the 24th international conference on machine learning, ICML ’07. ACM, New York, NY, USA, pp 1103–1110

  48. Yuan C (2011) Multi-task learning for bayesian matrix factorization. In: ICDM, pp 924–931

  49. Zhang Y, Yeung D-Y (2009) A convex formulation for learning task relationships in multi-task learning. In: Proceedings of the 26th conference on uncertainty in artificial intelligence (UAI 2009), pp 733–742

  50. Zhang Y, Yeung D-Y, Xu Q (2010) Probabilistic multi-task feature selection. In: Proceedings of the advances in neural information processing systems (NIPS 2010), pp 2559–2567

  51. Bouguila N, Ziou D (2011) A countably infinite mixture model for clustering and feature selection. Knowl Inf Syst, pp 1–20, ISSN 0219–1377. doi:10.1007/s10115-011-0467-4

  52. Zhou J, Yuan L, Liu J, Ye J (2011) A multi-task learning formulation for predicting disease progression. In: KDD, pp 814–822

Download references

Acknowledgments

This work has been supported by the National Science Foundation under Grant No. 0845951.

Author information

Authors and Affiliations

  1. EECS Department, University of Kansas, Lawrence, KS, USA

    Hongliang Fei & Jun Huan

Authors
  1. Hongliang Fei
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jun Huan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Hongliang Fei.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Fei, H., Huan, J. Structured feature selection and task relationship inference for multi-task learning. Knowl Inf Syst 35, 345–364 (2013). https://doi.org/10.1007/s10115-012-0543-4

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10115-012-0543-4

Keywords

Search

Navigation