A4 Article in conference proceedings
Quantifying Uncertainty in Machine Theory of Mind Across Time (2024)

Zhang, S., Wu, C., & Jokinen, J. P.P. (2024). Quantifying Uncertainty in Machine Theory of Mind Across Time. In J. Kasurinen, T. Päivärinta, & T. Vartiainen (Eds.), TKTP 2024 : Proceedings of the 41st Annual Doctoral Symposium of Computer Science (3776, pp. 151-156). RWTH Aachen. CEUR Workshop Proceedings. https://ceur-ws.org/Vol-3776/shortpaper14.pdf

JYU authors or editors

Publication details

All authors or editors: Zhang, Shanshan; Wu, Chuyang; Jokinen, Jussi P. P.

Parent publication: TKTP 2024 : Proceedings of the 41st Annual Doctoral Symposium of Computer Science

Parent publication editors: Kasurinen, Jussi; Päivärinta, Tero; Vartiainen, Tero

Place and date of conference: Vaasa, Finland, 10.-11.6.2024

Journal or series: CEUR Workshop Proceedings

eISSN: 1613-0073

Publication year: 2024

Volume: 3776

Pages range: 151-156

Number of pages in the book: 156

Publisher: RWTH Aachen

Publication country: Germany

Publication language: English

Persistent website address: https://ceur-ws.org/Vol-3776/shortpaper14.pdf

Publication open access: Openly available

Publication channel open access: Open Access channel

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/98364

Abstract

As intelligent interactive technologies advance, ensuring alignment with user preferences is critical. Machine theory of mind enablessystems to infer latent mental states from observed behaviors, similarly to humans. Currently, there is no formal mechanism for integrating multiple observations over time and quantifying the uncertainty of inferences as the function of accumulated evidence in a provably human-like way. This paper addresses the issue through Bayesian inference, proposing a model that maintains a posterior belief about mental states as a probability distribution, updated with observational data. The advantage of Bayesian statistics lies in the possibility of evaluating the certainty of these inferences. We validate the model’s human-like mental inference capabilities through an experiment.

Keywords: human-computer interaction; user interfaces; machine learning; modelling (representation)

Free keywords: human-computer interaction; machine theory of mind; mentalizing; uncertainty quantification

Contributing organizations

Ministry reporting: Yes

VIRTA submission year: 2024

Preliminary JUFO rating: 1