HUMANS DIFFER: SO SHOULD MODELS - Systematic Differences Call for Per-subject Modeling

Wolfgang Heidl, Stefan Thumfart, Christian Eitzinger

2012

Abstract

While machine learning is most often learning from humans, training data is still considered to originate from a uniform black box. Under this paradigm systematic differences in training provided by multiple subjects are translated into unavoidable modeling error. When trained on a per-subject basis those differences indeed translate to systematic differences in the resulting model structure. We feel that the goal of creating humanlike capabilities or behavior in artificial systems can only be achieved if the diversity of humans is adequately considered.

References

  1. Breiman, L. (2001). Random forests. Machine Learning, 45(1):5-32.
  2. Breiman, L., Friedman, J., Stone, C. J., and Olshen, R. A. (1993). Classification and Regression Trees. Chapman and Hall/CRC, Boca Raton, FL.
  3. Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associates, Mahwah, NJ, 2 edition.
  4. Donmez, P., Lebanon, G., and Balasubramanian, K. (2010). Unsupervised supervised learning i: Estimating classification and regression errors without labels. J. Mach. Learn. Res., 11:1323-1351.
  5. Drury, C. G. (1978). Integrating human factors models into statistical quality control. Human Factors: The Journal of the Human Factors and Ergonomics Society, 20(12):561-572.
  6. Eitzinger, C., Heidl, W., Lughofer, E., Raiser, S., Smith, J., Tahir, M., Sannen, D., and Van Brussel, H. (2009). Assessment of the influence of adaptive components in trainable surface inspection systems. Machine Vision and Applications, 21(5):613-626.
  7. Garrett, S. K., Melloy, B. J., and Gramopadhye, A. (2001). The effects of per-lot and per-item pacing on inspection performance. International Journal of Industrial Ergonomics, 27(5):291-302.
  8. Heidl, W., Thumfart, S., Lughofer, E., Eitzinger, C., and Klement, E. P. (2010). Classifier-based analysis of visual inspection: Gender differences in decisionmaking. In Proceedings of SMC2010, IEEE Conference on Systems, Man and Cybernetics, pages 113- 120.
  9. Heidl, W., Thumfart, S., Lughofer, E., Eitzinger, C., and Klement, E. P. (2011). Decision tree-based analysis suggests structural gender differences in visual inspection. In Proceedings of AIA2011, IASTED International Conference on Artificial Intelligence and Applications, pages 142-149.
  10. Jain, A. K., Prabhakar, S., and Hong, L. (1999). A multichannel approach to fingerprint classification. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 21(4):348-359.
  11. Moffat, S. D., Hampson, E., and Hatzipantelis, M. (1998). Navigation in a ”virtual” maze: Sex differences and correlation with psychometric measures of spatial ability in humans. Evolution and Human Behavior, 19(2):73-87.
  12. Peacock, A., Ke, X., and Wilkerson, M. (2004). Typing patterns: a key to user identification. Security Privacy, IEEE, 2(5):40-47.
  13. Ruggieri, S., Pedreschi, D., and Turini, F. (2010). Data mining for discrimination discovery. ACM Trans. Knowl. Discov. Data, 4(2):9:1-9:40.
  14. Stevens, R. and Soller, A. (2005). Machine learning models of problem space navigation: The influence of gender. Computer Science and Information Systems/ComSIS, 2(2):83-98.
  15. Thumfart, S., Jacobs, R. A., Lughofer, E., Eitzinger, C., Cornelissen, F. W., Groißböck, W., and Richter, R. (2011). Modelling human aesthetic perception of visual textures. Accepted for publication in ACM Trans. on Applied Perception.
  16. Yu, K., Wang, Y., and Tan, T. (2004). Writer identification using dynamic features. In Zhang, D. and Jain, A. K., editors, Biometric Authentication, volume 3072 of Lecture Notes in Computer Science, pages 1-8. Springer Berlin / Heidelberg.
  17. Zhao, W., Chellappa, R., Phillips, P. J., and Rosenfeld, A. (2003). Face recognition: A literature survey. ACM Comput. Surv., 35(4):399-458.
Download


Paper Citation


in Harvard Style

Heidl W., Thumfart S. and Eitzinger C. (2012). HUMANS DIFFER: SO SHOULD MODELS - Systematic Differences Call for Per-subject Modeling . In Proceedings of the 4th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-8425-95-9, pages 413-418. DOI: 10.5220/0003832904130418


in Bibtex Style

@conference{icaart12,
author={Wolfgang Heidl and Stefan Thumfart and Christian Eitzinger},
title={HUMANS DIFFER: SO SHOULD MODELS - Systematic Differences Call for Per-subject Modeling},
booktitle={Proceedings of the 4th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2012},
pages={413-418},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003832904130418},
isbn={978-989-8425-95-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 4th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - HUMANS DIFFER: SO SHOULD MODELS - Systematic Differences Call for Per-subject Modeling
SN - 978-989-8425-95-9
AU - Heidl W.
AU - Thumfart S.
AU - Eitzinger C.
PY - 2012
SP - 413
EP - 418
DO - 10.5220/0003832904130418