Dr. Tom Sterkenburg

I am currently principal investigator of the German Science Foundation-funded research project “The Epistemology of Statistical Learning Theory.” I hold a BSc in Artificial Intelligence (Amsterdam), a MSc in Logic (Amsterdam), a MSc in History and Philosophy of Science (Utrecht), and a PhD in Philosophy (joint at the University of Groningen and the CWI, the Dutch national research center for mathematics and computer science). In my PhD thesis (2018) I investigated the computability-theoretic approach to probabilistic “universal prediction” as a link between Carnap's inductive logic and modern approaches in machine learning.

Research Interests

My research is in the philosophy of induction and the epistemological foundations of machine learning.

Selected Publications

• On explaining the success of induction, The British Journal for the Philosophy of Science, forthcoming.
• On the truth-convergence of open-minded Bayesianism, The Review of Symbolic Logic, 2022. With Rianne de Heide.
• The no-free-lunch theorems of supervised learning, Synthese, 2021. With Peter Grünwald.
• The meta-inductive justification of induction, Episteme, 2020.
• The meta-inductive justification of induction: The pool of strategies, Philosophy of Science, 2019.
• Putnam’s diagonal argument and the impossibility of a universal learning machine, Erkenntnis, 2019.
• A generalized characterization of algorithmic probability, Theory of Computing Systems, 2017.
• Solomonoff prediction and Occam’s razor, Philosophy of Science, 2016.

Awards

Wolfgang Stegmüller Award for my PhD thesis