This page provides an overview of research at the intersection of the decision theory of Newcomb-like problems (including anthropics, naturalized induction, and other related topics) and artificial intelligence. Specifically, it lists papers, blog posts, etc. discussing the “implementation problem”: how do we implement decision theories in AIs? Conversely, what decision theories do different AI approaches implement?

• Albert and Heiner (2001): An Indirect-Evolution Approach to Newcomb’s Problem
• Meyer, Feldmaier and Shen (2016): Reinforcement Learning in Conflicting Environments for Autonomous Vehicles
• Caspar Oesterheld (2018): Doing what has worked well in the past leads to evidential decision theory
• Caspar Oesterheld (2018): Futarchy implements evidential decision theory
• Caspar Oesterheld (2018): The law of effect, randomization and Newcomb’s problem
• Caspar Oesterheld (2018): Goertzel’s GOLEM implements evidential decision theory applied to policy choice
• Caspar Oesterheld (Synthese 2019): Approval-directed agency and the decision theory of Newcomb-like problems
• James Bell, Linda Linsefors, Caspar Oesterheld, Joar Skalse (NeurIPS 2021): Reinforcement Learning in Newcomblike Environments.
• Caspar Oesterheld, Abram Demski and Vincent Conitzer (TARK 2023): A Theory of Bounded Inductive Rationality.
• Caspar Oesterheld, Johannes Treutlein, Roger Grosse, Vincent Conitzer, Jakob Foerster (NeurIPS 2023): Similarity-based cooperative equilibrium.
• Caspar Oesterheld, Emery Cooper, Miles Kodama, Linh Chi Nguyen, Ethan Perez: A dataset of questions on decision-theoretic reasoning in Newcomb-like problems.