Willem Röpke — accepted AI-conference papers

2026

Deep SPI: Safe Policy Improvement via World Models

Florent Delgrange, Raphaël Avalos, Willem Röpke

ICLR 2026poster

Safe policy improvement (SPI) offers theoretical control over policy updates, yet existing guarantees largely concern offline, tabular reinforcement learning (RL). We study SPI in general online settings, when combined with world model and representation learning. We develop a theoretical framework…

Cited by 0 Source ↗Scholar ↗

2023

Distributional Multi-Objective Decision Making

Willem Röpke, Conor F. Hayes, Patrick Mannion, Enda Howley, Ann Nowé, Diederik M. Roijers

IJCAI 2023poster

For effective decision support in scenarios with conflicting objectives, sets of potentially optimal solutions can be presented to the decision maker. We explore both what policies these sets should contain and how such sets can be computed efficiently. With this in mind, we take a distributional ap…

Cited by 6 Source ↗PDF ↗code ↗Scholar ↗