Seminars

Home Schools Computational Sciences Seminars

ABSTRACT: When there exists uncertainty, artificial intelligence (AI) machines are designed to optimize decisions based on expected outcomes. Given that expectations are built upon true facts about the objective environment the machines interact with, those facts can be encoded into AI models in the form of true objective probability functions. Consequently, many AI models incorporate probabilistic machine learning in which probability should be objectively interpreted. In this talk, we examine what probabilities are and demonstrate, through formal analysis, the conditions under which machines can learn the true objective probabilities—assuming they exist—and the conditions under which machines cannot learn them.