Deep multi-task learning for individuals origin–destination matrices estimation from census data

  • Mehdi KatranjiEmail author
  • Sami Kraiem
  • Laurent Moalic
  • Guilhem Sanmarty
  • Ghazaleh Khodabandelou
  • Alexandre Caminada
  • Fouad Hadj Selem


Rapid urbanization has made the estimation of the human mobility flows a substantial task for transportation and urban planners. Worker and student mobility flows are among the most weekly regular displacements and consequently generate road congestion issues. With urge of demands on efficient transport planning policies, estimating their commuting facilitates the decision-making processes for local authorities. Worker and student censuses often contain home location, work places and educational institutions. This paper proposes a novel approach to estimate individuals origin–destination matrices from census datasets. We use a multi-task neural network to learn a generic model providing the spatio-temporal estimations of commuters dynamic mobility flows on daily basis from static censuses. Multi-task learning aims at leveraging functional information incorporated in multiple tasks, which allows ameliorating the generalization performance within all the tasks. We first aggregate individuals household travel surveys and census databases with working and studying trips. The model learns the temporal distribution of displacements from these static sources and then it is applied on scholar and worker mobility sources to predict the temporal characteristics of commuters’ displacements (i.e. origin–destination matrices). Our method yields substantially more stable predictions in terms of accuracy and results in a significant error rate control in comparison to single task learning.


Origin–destination matrix updating/estimating Regular mobility pattern Multi-task learning 



The authors would like to thank both CEREMA (provider), ADISP-CMH (distributor) for the HTS datasets (ADISP, 2018), the ANR for granting the project Norm-Atis under Grant ANR-13-TDMO-07 and Orange Fluxvision. (Orange, 2018).

Author Contributions

MK coding, data analysis, drafting manuscript, experimentation, mathematical analyses, study design. SK data acquisition, data analysis, manuscript correction. LM manuscript correction. GS data acquisition, data analysis, manuscript correction. GK data analysis, drafting manuscript, manuscript correction. AC manuscript correction. FH-S supervising


  1. Abdulnabi AH, Wang G, Lu J, Jia K (2015) Multi-task CNN model for attribute prediction multi-task cnn model for attribute prediction. IEEE Trans Multimed 17(11):1949–1959CrossRefGoogle Scholar
  2. Abu-Mostafa YS (1990) Learning from hints in neural networks. J Complex 6(2):192–198MathSciNetCrossRefGoogle Scholar
  3. ADISP (2018) Archives de données issues de la statistique publique.
  4. Ahmed A, Aly M, Das A, Smola AJ, Anastasakos T (2012) Web-scale multi-task feature selection for behavioral targeting. In: Proceedings of the 21st ACM international conference on Information and knowledge management, pp 1737–1741Google Scholar
  5. Alamgir M, Grosse-Wentrup M, Altun Y (2010) Multitask learning for brain-computer interfaces. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp 17–24Google Scholar
  6. Arai A, Shibasaki R (2013) Estimation of human mobility patterns and attributes analyzing anonymized mobile phone CDR: developing real-time census from crowds of greater dhaka. In: Agile PhD schoolGoogle Scholar
  7. Arai A, Witayangkurn A, Kanasugi H, Horanont T, Shao X, Shibasaki R (2014) Understanding user attributes from calling behavior: exploring call detail records through field observations. In: Proceedings of the 12th international conference on advances in mobile computing and multimedia, pp 95–104Google Scholar
  8. Bachir D, Gauthier V, El Yacoubi M, Khodabandelou G (2017) Using mobile phone data analysis for the estimation of daily urban dynamics. In: 2017 IEEE 20th international conference on intelligent transportation systems (ITSC), pp 626–632Google Scholar
  9. Bachir D, Khodabandelou G, Gauthier V, El Yacoubi M, Vachon E (2018) Combining bayesian inference and clustering for transport mode detection from sparse and noisy geolocation data. In: Joint European conference on machine learning and knowledge discovery in databases, pp 569–584CrossRefGoogle Scholar
  10. Bachir D, Khodabandelou G, Gauthier V, El Yacoubi M, Puchinger J (2019) Inferring dynamic origin–destination flows by transport mode using mobile phone data. Trans Res Part C Emerg Technol 101:254–275CrossRefGoogle Scholar
  11. Baxter J (2000) A model of inductive bias learning. J Artif Intell Res 12:149–198MathSciNetCrossRefGoogle Scholar
  12. Caruana R (1997) Multitask learning. Mach Learn 28(1):41–75MathSciNetCrossRefGoogle Scholar
  13. Ceci M, Corizzo R, Fumarola F, Malerba D, Rashkovska A (2017) Predictive modeling of pv energy production: how to set up the learning task for a better prediction? IEEE Trans Ind Inform 13(3):956–966CrossRefGoogle Scholar
  14. CEREMA (2012) Urban mobility in france. Main lessons learnt from the years 2000–2010.
  15. Cirstea, R-G, Micu D-V, Muresan G-M, Guo C, Yang B (2018) Correlated time series forecasting using multi-task deep neural networks. In: Proceedings of the 27th acm international conference on information and knowledge management, pp 1527–1530Google Scholar
  16. Collobert R, Weston J (2008) A unified architecture for natural language processing: Deep neural networks with multitask learning. In: Proceedings of the 25th international conference on machine learning, pp 160–167Google Scholar
  17. Côté M-A, Larochelle H (2016) An infinite restricted boltzmann machine. Neural Comput 28(7):1265–1288MathSciNetCrossRefGoogle Scholar
  18. Ding S, Jia W, Su C, Zhang L, Shi Z (2008) Neural network research progress and applications in forecast. In: International symposium on neural networks, pp 783–793Google Scholar
  19. Gaudette L, Japkowicz N (2009) Evaluation methods for ordinal classification. In: Canadian conference on artificial intelligence, pp 207–210CrossRefGoogle Scholar
  20. Ghosn J, Bengio Y (1997) Multi-task learning for stock selection. In: Advances in neural information processing systems, pp 946–952Google Scholar
  21. Giannotti F, Nanni M, Pedreschi D, Pinelli F, Renso C, Rinzivillo S, Trasarti R (2011) Unveiling the complexity of human mobility by querying and mining massive trajectory data. VLDB J Int J Very Large Data Bases 20(5):695–719CrossRefGoogle Scholar
  22. Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp 249–256Google Scholar
  23. Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Proceedings of the fourteenth international conference on artificial intelligence and statistics, pp 315–323Google Scholar
  24. Gonzalez MC, Hidalgo CA, Barabasi A-L (2008) Understanding individual human mobility patterns. Nature 453(7196):779CrossRefGoogle Scholar
  25. GraphHopper (2018) Graphhopper directions api with route optimization.
  26. He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on computer vision, pp 1026–1034Google Scholar
  27. Hinton GE, Salakhutdinov RR (2006) Reducing the dimensionality of data with neural networks. Science 313(5786):504–507MathSciNetCrossRefGoogle Scholar
  28. Hinton GE, Srivastava N, Krizhevsky A, Sutskever I, Salakhutdinov RR (2012) Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580
  29. Hopfield JJ (1987) Neural networks and physical systems with emergent collective computational abilities. In: Spin glass theory and beyond: an introduction to the replica method and its applications. World Scientific, pp 411–415Google Scholar
  30. Hornik K (1991) Approximation capabilities of multilayer feedforward networks. Neural Netw 4(2):251–257MathSciNetCrossRefGoogle Scholar
  31. Hu Q, Wu Z, Richmond K, Yamagishi J, Stylianou Y, Maia R (2015) Fusion of multiple parameterisations for dnn-based sinusoidal speech synthesis with multi-task learning. In: Sixteenth annual conference of the international speech communication associationGoogle Scholar
  32. INSEE (2018) National institute of statistics and economic studies.
  33. Katranji M, Thuillier E, Kraiem S, Moalic L, Selem FH (2016) Mobility data disaggregation: A transfer learning approach. In: 2016 IEEE 19th international conference on intelligent transportation systems (ITSC), pp 1672–1677Google Scholar
  34. Katranji M, Moalic L, Sanmarty G, Kraiem S, Caminada A, Hadj Selem F (2018) Mixed-variate restricted boltzmann machines for the inference of origin–destination matrices. In: TRB (transportation research board) annual meeting (2018)Google Scholar
  35. Katranji M, Sanmarty G, Moalic L, Kraiem S, Caminada A, Selem FH (2018) Rnn encoder-decoder for the inference of regular human mobility patterns. In: 2018 international joint conference on neural networks (IJCNN), pp 1–9Google Scholar
  36. Kendall A, Gal Y, Cipolla R (2018) Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 7482–7491Google Scholar
  37. Khodabandelou G, Gauthier V, El-Yacoubi M, Fiore M (2016) Population estimation from mobile network traffic metadata. In: 2016 IEEE 17th international symposium on a world of wireless, mobile and multimedia networks (WOWMOM), pp 1–9Google Scholar
  38. Khodabandelou G, Gauthier V, Fiore M, El Yacoubi MA (2018) Estimation of static and dynamic urban populations with mobile network metadata. IEEE Trans Mob Comput 18:2034–2047CrossRefGoogle Scholar
  39. Kingma DP, Welling M (2013) Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114
  40. Kingma D, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980
  41. Lenormand M, Picornell M, Cantú-Ros OG, Tugores A, Louail T, Herranz R, Ramasco JJ (2014) Cross-checking different sources of mobility information. PLoS ONE 9(8):e105184CrossRefGoogle Scholar
  42. Lenormand M, Louail T, Cantú-Ros OG, Picornell M, Herranz R, Arias JM, Ramasco JJ (2015) Influence of sociodemographic characteristics on human mobility. Sci Rep 5:10075CrossRefGoogle Scholar
  43. Long M, Wang J (2015) Learning multiple tasks with deep relationship networks. arXiv preprint arXiv:1506.02117
  44. Louf R, Barthelemy M (2014) How congestion shapes cities: from mobility patterns to scaling. Sci Rep 4:5561CrossRefGoogle Scholar
  45. Louf R, Roth C, Barthelemy M (2014) Scaling in transportation networks. PLoS ONE 9(7):e102007CrossRefGoogle Scholar
  46. Navitia (2018) The open api for building cool stuff with transport data.
  47. Pan SJ, Yang Q (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22(10):1345–1359CrossRefGoogle Scholar
  48. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O (2011) Scikit-learn: Machine learning in python. J Mach Learn Res 12(Oct):2825–2830MathSciNetzbMATHGoogle Scholar
  49. Prechelt L (1998) Early stopping—but when? In: Montavon G, Orr GB, Müller K-R (eds) Neural networks: tricks of the trade. Springer, Berlin, pp 53–67Google Scholar
  50. Puniyani K, Kim S, Xing EP (2010) Multi-population GWA mapping via multi-task regularized regression. Bioinformatics 26(12):i208–i216CrossRefGoogle Scholar
  51. Ruder S (2017) An overview of multi-task learning in deep neural networks. arXiv preprint arXiv:1706.05098
  52. Rumelhart DE, Hinton GE, Williams RJ et al (1988) Learning representations by back-propagating errors. Cogn Model 5(3):1zbMATHGoogle Scholar
  53. Smoreda Z, Olteanu-Raimond A-M, Couronné T et al (2013) Spatiotemporal data from mobile phones for personal mobility assessment. Trans Surv Methods Best Pract Decis Mak 41:745–767CrossRefGoogle Scholar
  54. Song X, Kanasugi H, Shibasaki R (2016) Deeptransport: prediction and simulation of human mobility and transportation mode at a citywide level. In: IJCAI, vol 16, pp 2618–2624Google Scholar
  55. Toqué F, Côme E, El Mahrsi MK, Oukhellou L (2016) Forecasting dynamic public transport origin–destination matrices with long-short term memory recurrent neural networks. In: 2016 IEEE 19th international conference on intelligent transportation systems (ITSC), pp 1071–1076Google Scholar
  56. Tran T, Phung D, Venkatesh S (2011) Mixed-variate restricted boltzmann machines. In: Asian conference on machine learning, pp 213–229Google Scholar
  57. Wachowicz M, Ong R, Renso C, Nanni M (2011) Finding moving flock patterns among pedestrians through collective coherence. Int J Geogr Inf Sci 25(11):1849–1864CrossRefGoogle Scholar
  58. Willumsen LG (1978) Estimation of an O–D matrix from traffic counts? a review. Institute of Transport Studies, University of Leeds.
  59. Wu Z, Valentini-Botinhao C, Watts O, King S (2015) Deep neural networks employing multi-task learning and stacked bottleneck features for speech synthesis. In: 2015 IEEE international conference on acoustics, speech and signal processing (ICASSP), pp 4460–4464Google Scholar
  60. Xue Y, Liao X, Carin L, Krishnapuram B (2007) Multi-task learning for classification with dirichlet process priors. J Mach Learn Res 8(Jan):35–63MathSciNetzbMATHGoogle Scholar
  61. Yang Y, Hospedales T (2016) Deep multi-task representation learning: a tensor factorisation approach. arXiv preprint arXiv:1605.06391
  62. Yim J, Jung H, Yoo B, Choi C, Park D, Kim J (2015) Rotating your face using multi-task deep neural network. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 676–684Google Scholar
  63. Zhang Y, Yeung D-Y (2012) A convex formulation for learning task relationships in multi-task learning. arXiv preprint arXiv:1203.3536
  64. Zhang Y, Yeung D-Y (2014) A regularization approach to learning task relationships in multitask learning. ACM Trans Knowl Discov Data (TKDD) 8(3):12Google Scholar
  65. Zhang Y, Yang Q (2017) A survey on multi-task learning. arXiv preprint arXiv:1707.08114
  66. Zhang K, Gray JW, Parvin B (2010) Sparse multitask regression for identifying common mechanism of response to therapeutic targets. Bioinformatics 26(12):i97–i105CrossRefGoogle Scholar
  67. Zhang D, Huang J, Li Y, Zhang F, Xu C, He T (2014) Exploring human mobility with multi-source data at extremely large metropolitan scales. In: Proceedings of the 20th annual international conference on mobile computing and networking, pp 201–212Google Scholar
  68. Zhang D, Zhao J, Zhang F, He T (2015) comobile: real-time human mobility modeling at urban scale using multi-view learning. In: Proceedings of the 23rd sigspatial international conference on advances in geographic information systems, p 40Google Scholar
  69. Zheng Y, Capra L, Wolfson O, Yang H (2014) Urban computing: concepts, methodologies, and applications. ACM Trans Intell Syst Technol (TIST) 5(3):38Google Scholar
  70. Zhou J, Chen J, Ye J (2012) Multi-task learning: theory, algorithms, and applications. SDM tutorialsGoogle Scholar

Copyright information

© The Author(s), under exclusive licence to Springer Science+Business Media LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.VEDECOMVersaillesFrance
  2. 2.IRIMAS EA 7499University of Haute-AlsaceMulhouseFrance
  3. 3.University of Nice Sophia AntipolisNiceFrance

Personalised recommendations