Continuing
cognitive information fusion: interpretable inference in the presence of uncertain uncertainty and unknown unknowns
Key Points
- DAIRNet Theme: Autonomous Processing and Reasoning
- University of Technology Sydney
- Point of contact: A/Prof Adrian Bishop
This project addresses methodology, theory, and technologies associated with interpretable inference and decision making in the presence of “uncertain uncertainty” and “unknown unknowns”. By “uncertain uncertainty” we refer to those complex situations in which the information sources are not finely detailed enough to specify low-level probabilistic uncertainty, or the correlation between observed data points is unknown. By “unknown unknowns” we refer to unexpected or unforeseeable conditions, which pose a potentially great risk simply because they cannot be anticipated based on past experience.
