Marc P. Deisenroth

Alumni

Empirical Inference

Publications

4 results (View BibTeX file of all listed publications)

2017

Model-based Contextual Policy Search for Data-Efficient Generalization of Robot Skills

Kupcsik, A., Deisenroth, M., Peters, J., Ai Poh, L., Vadakkepat, V., Neumann, G.

Artificial Intelligence, 247, pages: 415-439, 2017, Special Issue on AI and Robotics (article)

link (url) DOI Project Page [BibTex]

2017

Kupcsik, A., Deisenroth, M., Peters, J., Ai Poh, L., Vadakkepat, V., Neumann, G. Model-based Contextual Policy Search for Data-Efficient Generalization of Robot Skills Artificial Intelligence, 247, pages: 415-439, 2017, Special Issue on AI and Robotics (article)

link (url) DOI Project Page [BibTex]

2013

Probabilistic movement modeling for intention inference in human-robot interaction

Wang, Z., Mülling, K., Deisenroth, M., Ben Amor, H., Vogt, D., Schölkopf, B., Peters, J.

International Journal of Robotics Research, 32(7):841-858, 2013 (article)

PDF DOI [BibTex]

2013

Wang, Z., Mülling, K., Deisenroth, M., Ben Amor, H., Vogt, D., Schölkopf, B., Peters, J. Probabilistic movement modeling for intention inference in human-robot interaction International Journal of Robotics Research, 32(7):841-858, 2013 (article)

PDF DOI [BibTex]

2012

Probabilistic Modeling of Human Movements for Intention Inference

Wang, Z., Deisenroth, M., Ben Amor, H., Vogt, D., Schölkopf, B., Peters, J.

In Proceedings of Robotics: Science and Systems VIII, pages: 8, R:SS, 2012 (inproceedings)

Abstract

Inference of human intention may be an essential step towards understanding human actions [21] and is hence important for realizing efficient human-robot interaction. In this paper, we propose the Intention-Driven Dynamics Model (IDDM), a latent variable model for inferring unknown human intentions. We train the model based on observed human behaviors/actions and we introduce an approximate inference algorithm to efficiently infer the human’s intention from an ongoing action. We verify the feasibility of the IDDM in two scenarios, i.e., target inference in robot table tennis and action recognition for interactive humanoid robots. In both tasks, the IDDM achieves substantial improvements over state-of-the-art regression and classification.

PDF link (url) [BibTex]