Decision-Making under Uncertainty

Welcome to the home page of the Decision-Making under Uncertainty Multi-University Research Initiative: a multidisciplinary research effort that brings together sixteen principal investigators from Stanford University, the University of California (Berkeley, Davis, Irvine, Los Angeles) and the University of Illinois at Urbana-Champaign.

The high level objective of this research effort is to:

develop methods and techniques for dealing with uncertainty in all its manifestations; and
build research prototypes of decision aids that incorporate these methods.

The decision aids will ultimately help decision makers in developing plans and contingencies that are robust (degrade gracefully), are easy to use, and have the greatest likelihood of success.

Research Concentration Areas: The PIs involved in this research effort are addressing a variety of research topics which include but are not limited to the following:

Basic methods based on probability, decision and utility theories
Techniques for bounding the effect of missing and/or incorrect information
Trading time and space resources with certainty
Fusing uncertain information of different kinds
Real-time inference algorithms
Hybrid dynamical systems
Machine learning algorithms
Causal reasoning
Problem structuring for optimal understanding by human decision makers
Analysis and evaluation of courses of actions
Cognitive aspects of interaction between human decision makers
Methods for evaluating the effectiveness of automated decision aids

Impact: In the long term, this research effort has the potential to fundamentally change the decision-making capabilities of both humans and machines. The ability to deal with uncertain information in a straightforward and natural way is expected to: improve the quality of planning; enable more rational responses to unexpected events; provide new levels of situation awareness; and allow a better understanding of available options. These enhancements to the decision making process will enable people and machines to make better decisions in less time and with lower costs.

muri@gubbio.cs.berkeley.edu