My name is Caspar Oesterheld (anglicized pronunciation) and I am a Computer Science PhD student at Duke University, supervised by Vincent Conitzer. My interests lie at the intersection of game and decision theory as well as artifial intelligence and theoretical computer science. In particular, I wonder how intelligent agents should reason when
Cartesian barriers break down. For example, how should programs play a game if they can read each other's source code? While I'm also doing some semi-philosophical work, being a computer scientist by training, I (nowadays) primarily work on these questions from a technical perspective. An example is my paper titled Robust Program Equilibrium, published in Theory and Decision.
- James Bell*, Linda Linsefors*, Caspar Oesterheld*, and Joar Skalse*: "Reinforcement Learning in Newcomblike Environments".
- Scott Emmons, Caspar Oesterheld, Andrew Critch, Vincent Conitzer, and Stuart Russell: "Symmetry, Equilibria, and Robustness in Common-Payoff Games". Presented at GAIW 2021.
- Caspar Oesterheld, Abram Demski, and Vincent Conitzer: "A theory of bounded inductive rationality". Presented at OptLearnMAS 2021.
- Johannes Treutlein, Michael Dennis, Caspar Oesterheld, and Jakob Foerster: "A New Formalism, Method and Open Issues for Zero-Shot Coordination". ICML 2021.
- Caspar Oesterheld and Vincent Conitzer: "Safe Pareto improvements for delegated game playing". AAMAS 2021. Also presented at the Cooperative AI Workshop at NeurIPS 2020.
- Caspar Oesterheld and Vincent Conitzer: "Decision Scoring Rules". WINE 2020. Also (to be) presented at CMID 2020, GAMES 2020/1, DIMACS Forecasting 2021, GAIW 2021, COMSOC 2021. [Extended version, Poster, talk at CMID 2020]
- Caspar Oesterheld and Vincent Conitzer: "Minimum-regret contracts for principal-expert problems". WINE 2020.
- Caspar Oesterheld and Vincent Conitzer: "Extracting Money from Causal Decision Theorists". Forthcoming in The Philosophical Quarterly. Also presented at the IJCAI-PRICAI 2020 AI Safety workshop and to be presented at GAMES 2020/1.
- Caspar Oesterheld: "Approval-directed agency and the decision theory of Newcomb-like problems". To appear in: Synthese Special Issue on Decision Theory and the Future of Artificial Intelligence. Shared second prize winner of the AI alignment prize.
- William MacAskill, Aron Vallinder, Caspar Oesterheld, Carl Shulman, and Johannes Treutlein: "The Evidentialist's Wager". Forthcoming in The Journal of Philosophy.
- Caspar Oesterheld: Robust Program Equilibrium. Theory and Decision, Volume 86, pages 143–159, February 2019.
- Caspar Oesterheld: Formalizing preference utilitarianism in physical world models. Synthese, Volume 193, pages 2747–2759, September 2016.
TeachingIn 2020, I received the Duke CS Department's Outstanding TA award.
- Fall 2019: TA for CS 230 – Discrete Mathematics for Computer Science – Instructor: Prof. Bruce Donald
- Spring 2020: TA for CS 323 – Computational Microeconomics – Instructor: Prof. Vincent Conitzer
- Fall 2020: TA for CS 590.7 – Computational Microeconomics – Instructor: Prof. Vincent Conitzer