A Complete Criterion for Value of Information in Soluble Influence Diagrams
This provides a foundational tool for researchers and practitioners in AI safety and fairness, though it appears incremental as it builds on existing influence diagram methods.
The paper tackles the problem of analyzing safety and fairness in AI systems by establishing the first complete graphical criterion for value of information in influence diagrams with multiple decisions, using techniques like ID homomorphisms and Tree of Systems to prove properties.
Influence diagrams have recently been used to analyse the safety and fairness properties of AI systems. A key building block for this analysis is a graphical criterion for value of information (VoI). This paper establishes the first complete graphical criterion for VoI in influence diagrams with multiple decisions. Along the way, we establish two important techniques for proving properties of multi-decision influence diagrams: ID homomorphisms are structure-preserving transformations of influence diagrams, while a Tree of Systems is collection of paths that captures how information and control can flow in an influence diagram.