Reasoning about Probabilities in Dynamic Systems using Goal Regression

Vaishak Belle, Hector Levesque
Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, PMLR R11:292-301, 2013.

Abstract

Reasoning about degrees of belief in uncertain dy- namic worlds is fundamental to many applications, such as robotics and planning, where actions mod- ify state properties and sensors provide measurements, both of which are prone to noise. With the exception of limited cases such as Gaussian processes over lin- ear phenomena, belief state evolution can be complex and hard to reason with in a general way. This pa- per proposes a framework with new results that allows the reduction of subjective probabilities after sensing and acting, both in discrete and continuous domains, to questions about the initial state only. We build on an expressive probabilistic first-order logical ac- count by Bacchus, Halpern and Levesque, resulting in a methodology that, in principle, can be coupled with a variety of existing inference solutions.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR11-belle13a, title = {Reasoning about Probabilities in Dynamic Systems using Goal Regression}, author = {Belle, Vaishak and Levesque, Hector}, booktitle = {Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence}, pages = {292--301}, year = {2013}, editor = {Nicholson, Ann and Smyth, Padhraic}, volume = {R11}, series = {Proceedings of Machine Learning Research}, month = {12--14 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r11/main/assets/belle13a/belle13a.pdf}, url = {https://proceedings.mlr.press/r11/belle13a.html}, abstract = {Reasoning about degrees of belief in uncertain dy- namic worlds is fundamental to many applications, such as robotics and planning, where actions mod- ify state properties and sensors provide measurements, both of which are prone to noise. With the exception of limited cases such as Gaussian processes over lin- ear phenomena, belief state evolution can be complex and hard to reason with in a general way. This pa- per proposes a framework with new results that allows the reduction of subjective probabilities after sensing and acting, both in discrete and continuous domains, to questions about the initial state only. We build on an expressive probabilistic first-order logical ac- count by Bacchus, Halpern and Levesque, resulting in a methodology that, in principle, can be coupled with a variety of existing inference solutions.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Reasoning about Probabilities in Dynamic Systems using Goal Regression %A Vaishak Belle %A Hector Levesque %B Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2013 %E Ann Nicholson %E Padhraic Smyth %F pmlr-vR11-belle13a %I PMLR %P 292--301 %U https://proceedings.mlr.press/r11/belle13a.html %V R11 %X Reasoning about degrees of belief in uncertain dy- namic worlds is fundamental to many applications, such as robotics and planning, where actions mod- ify state properties and sensors provide measurements, both of which are prone to noise. With the exception of limited cases such as Gaussian processes over lin- ear phenomena, belief state evolution can be complex and hard to reason with in a general way. This pa- per proposes a framework with new results that allows the reduction of subjective probabilities after sensing and acting, both in discrete and continuous domains, to questions about the initial state only. We build on an expressive probabilistic first-order logical ac- count by Bacchus, Halpern and Levesque, resulting in a methodology that, in principle, can be coupled with a variety of existing inference solutions. %Z Reissued by PMLR on 04 October 2026.
APA
Belle, V. & Levesque, H.. (2013). Reasoning about Probabilities in Dynamic Systems using Goal Regression. Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R11:292-301 Available from https://proceedings.mlr.press/r11/belle13a.html. Reissued by PMLR on 04 October 2026.

Related Material