Belief Functions

Belief functions, also called mass functions or basic probability assignments (BPA), provide a mathematical framework for representing uncertainty and combining evidence from multiple sources. Unlike traditional probability theory which assigns probabilities to individual outcomes, belief functions assign belief mass to sets of outcomes, capturing both the available evidence and the absence of information.

Belief functions are the foundation of Dempster-Shafer theory and have become increasingly important in information fusion applications where multiple sources provide conflicting or incomplete information about a proposition.

Key papers

• Belief Evolution Network-based Probability Transformation and Fusion — introduces Belief Evolution Network framework for probability transformation

Related topics

• Dempster-Shafer Theory (foundational theory)
• Information Fusion (primary application area)
• Uncertainty Quantification (broader context)