Predicting the behavior of interacting humans by fusing data from multiple sources

Erik J. Schlicht, Ritchie Lee, David H. Wolpert, Mykel J. Kochenderfer, Brendan Tracey
Proceedings of the 28th Conference on Uncertainty in Artificial Intelligence, PMLR R10:755-763, 2012.

Abstract

Multi-fidelity methods combine inexpensive low-fidelity simulations with costly but highfidelity simulations to produce an accurate model of a system of interest at minimal cost. They have proven useful in modeling physical systems and have been applied to engineering problems such as wing-design optimization. During human-in-the-loop experimentation, it has become increasingly common to use online platforms, like Mechanical Turk, to run low-fidelity experiments to gather human performance data in an efficient manner. One concern with these experiments is that the results obtained from the online environment generalize poorly to the actual domain of interest. To address this limitation, we extend traditional multi-fidelity approaches to allow us to combine fewer data points from high-fidelity human-in-the-loop experiments with plentiful but less accurate data from low-fidelity experiments to produce accurate models of how humans interact. We present both model-based and model-free methods, and summarize the predictive performance of each method under dierent conditions.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR10-schlicht12a, title = {Predicting the behavior of interacting humans by fusing data from multiple sources}, author = {Schlicht, Erik J. and Lee, Ritchie and Wolpert, David H. and Kochenderfer, Mykel J. and Tracey, Brendan}, booktitle = {Proceedings of the 28th Conference on Uncertainty in Artificial Intelligence}, pages = {755--763}, year = {2012}, editor = {de Freitas, Nando and Murphy, Kevin}, volume = {R10}, series = {Proceedings of Machine Learning Research}, month = {14--18 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r10/main/assets/schlicht12a/schlicht12a.pdf}, url = {https://proceedings.mlr.press/r10/schlicht12a.html}, abstract = {Multi-fidelity methods combine inexpensive low-fidelity simulations with costly but highfidelity simulations to produce an accurate model of a system of interest at minimal cost. They have proven useful in modeling physical systems and have been applied to engineering problems such as wing-design optimization. During human-in-the-loop experimentation, it has become increasingly common to use online platforms, like Mechanical Turk, to run low-fidelity experiments to gather human performance data in an efficient manner. One concern with these experiments is that the results obtained from the online environment generalize poorly to the actual domain of interest. To address this limitation, we extend traditional multi-fidelity approaches to allow us to combine fewer data points from high-fidelity human-in-the-loop experiments with plentiful but less accurate data from low-fidelity experiments to produce accurate models of how humans interact. We present both model-based and model-free methods, and summarize the predictive performance of each method under dierent conditions.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Predicting the behavior of interacting humans by fusing data from multiple sources %A Erik J. Schlicht %A Ritchie Lee %A David H. Wolpert %A Mykel J. Kochenderfer %A Brendan Tracey %B Proceedings of the 28th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2012 %E Nando de Freitas %E Kevin Murphy %F pmlr-vR10-schlicht12a %I PMLR %P 755--763 %U https://proceedings.mlr.press/r10/schlicht12a.html %V R10 %X Multi-fidelity methods combine inexpensive low-fidelity simulations with costly but highfidelity simulations to produce an accurate model of a system of interest at minimal cost. They have proven useful in modeling physical systems and have been applied to engineering problems such as wing-design optimization. During human-in-the-loop experimentation, it has become increasingly common to use online platforms, like Mechanical Turk, to run low-fidelity experiments to gather human performance data in an efficient manner. One concern with these experiments is that the results obtained from the online environment generalize poorly to the actual domain of interest. To address this limitation, we extend traditional multi-fidelity approaches to allow us to combine fewer data points from high-fidelity human-in-the-loop experiments with plentiful but less accurate data from low-fidelity experiments to produce accurate models of how humans interact. We present both model-based and model-free methods, and summarize the predictive performance of each method under dierent conditions. %Z Reissued by PMLR on 04 October 2026.
APA
Schlicht, E.J., Lee, R., Wolpert, D.H., Kochenderfer, M.J. & Tracey, B.. (2012). Predicting the behavior of interacting humans by fusing data from multiple sources. Proceedings of the 28th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R10:755-763 Available from https://proceedings.mlr.press/r10/schlicht12a.html. Reissued by PMLR on 04 October 2026.

Related Material