Uncertain observation times

Authors:
Shaunak Chatterjee;Stuart Russell
Affiliations:
Computer Science Division, University of California, Berkeley, CA;Computer Science Division, University of California, Berkeley, CA
Venue:
SUM'12 Proceedings of the 6th international conference on Scalable Uncertainty Management
Year:
2012

Citing 3
Cited 0

Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
Learning for control from multiple demonstrations

Proceedings of the 25th international conference on Machine learning
Continuous time bayesian networks

UAI'02 Proceedings of the Eighteenth conference on Uncertainty in artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Standard temporal models assume that observation times are correct, whereas in many real-world settings (particularly those involving human data entry) noisy time stamps are quite common. Serious problems arise when these time stamps are taken literally. This paper introduces a modeling framework for handling uncertainty in observation times and describes inference algorithms that, under certain reasonable assumptions about the nature of time-stamp errors, have linear time complexity.