Research

My research lies at the intersection of formal epistemology, decision theory, statistics, and computer science. Much of my work concerns rational belief and choice under uncertainty: how agents should represent incomplete or ambiguous evidence, learn from new information, and make decisions when the available evidence does not uniquely determine what to believe or do.

Rational belief and decision-making under uncertainty

My current research investigates the foundations and limitations of Bayesian and imprecise-probability approaches to rational belief and decision-making. I am particularly interested in cases of evidential and probabilistic underdetermination, where evidence constrains rational credence without determining a unique probability distribution.

One strand of this work examines the role of prior beliefs in Bayesian learning. In settings where new evidence determines only some features of a posterior distribution, prior-sensitive updating rules can preserve aspects of the agent's prior that are not fixed by the new evidence. My work investigates when this inductive bias is epistemically or decision-theoretically significant, including cases in which truthful probabilistic information fails to improve, or can even worsen, belief accuracy and decision quality.

From October 2026, I will develop these ideas further in my research project Structural Permissivism: Stake-Sensitive Rational Belief under Evidential Underdetermination at Paris Lodron University of Salzburg. The project asks how rational agents should act when several probability distributions remain compatible with the evidence. It develops an account on which structural features of decision problems can play a principled role in determining which beliefs should guide action, without relaxing evidential constraints on rational belief.

Imprecise probability and rational choice

A related strand of my research concerns imprecise probability models. These approaches can represent evidential underdetermination without requiring an agent to adopt a single precise probability distribution, but they raise difficult questions about action guidance. I am interested in when several actions should count as genuinely permissible and when underdetermination instead provides grounds for rational suspension of choice.

My current work connects these questions with reason-based theories of rational choice. The aim is to develop a framework in which structural features of a decision problem can provide reasons for selecting among evidentially admissible representations — or, where such reasons are absent, for withholding such a selection.

Game theory and human–AI interaction

My broader research also includes game theory and the study of interaction between human and artificial agents. I have worked on team reasoning, coordination, and the foundations of strategic choice, and I am interested in how these ideas extend to settings in which humans interact with artificial learning agents. More generally, I am interested in whether formal models of rational belief and choice can help explain coordination, trust, and strategic behaviour in mixed human–AI environments.

Publications

1. Ambiguity, Coherence and Performance (with William Peden, Daniele Tortoli and Francesco De Pretis), Economics and Philosophy, 2025. 

2. Ambiguous Decisions in Bayesianism and Imprecise Probability (with William Peden and Francesco De Pretis), British Journal for Philosophy of Science Short Reads, 2024. 

3. A Comparison of Imprecise Bayesianism and Dempster-Shafer Theory for Automated Decisions under Ambiguity (with William Peden, Daniele Tortoli and Francesco De Pretis), Journal of Logic and Computation, 2024. 

4. The Ambiguity Dilemma for Imprecise Bayesians (with William Peden and Francesco De Pretis), The British Journal for the Philosophy of Science, forthcoming.

5. Making Decisions with Evidential Probability and Objective Bayesian Calibration Inductive Logics (with William Peden and Francesco De Pretis), International Journal of Approximate reasoning, 2023, 162. 

6. Incentives for Research Effort: an Evolutionary Model of Publication Markets with Double Blind and Open Review (with Francesco De Pretis, William Peden, Daniele Tortoli and Barbara Osimani), Computational Economics, 2023, 61(1). 

7. Team Reasoning without a Hive Mind (with Jurgis Karpus), Research in Economics, 2021, 75(4). 

8. A Battle in the Statistics Wars. A Simulation-Based Comparison of Bayesian, Frequentist and Williamsonian Methodologies (with William Peden and Francesco De Pretis), Synthese, 2021, 199.

9. Game Theory and Rational Reasoning (with Jurgis Karpus), The Routledge Handbook of Philosophy of Economics, Routledge, 2021. 

10. Team Reasoning and a Measure of Mutual Advantage in Games (with Jurgis Karpus), Economics and Philosophy, 2018, 34(1)

11. The Ant Trap: Rebuilding the Foundations of the Social Sciences, Brian Epstein. Oxford University Press 2015, viii + 298 pages, Economics and Philosophy, 2016, 32(2)

12. Shifting Systems of Knowledge as a Process of Modernization of the Political Philosophy: the Logic of Inversion between the Theoretical Model and Lifeworld, Logos, 2011, 66 (in Lithuanian) 

13. Rawlsian "Justice" and the Evolutionary Theory of Games: Cultural Evolution and the Origins of the Natural Maximin Rule, Problemos, 2011, 80

Work-in-progress papers

• Interactive decision-making for real (with Jurgis Karpus)

• Prior and prejudice (with William Peden)

• Rational coordination and emergent incentives

• The Limits of Bayesian Learning

• Dependency equilibria: extending Nash equilibria to entangled belief systems (with Wolfgang Spohn and Gerard Rothfus) 

• Belief inertia and communication with ambiguous signals