Below is a list of my publications, with links to conference/journal papers, workshop papers, preprints, supplements, and code where available. Please feel free to get in touch if you have any questions about my research.
Conference and Journal Papers
Learning The Minimum Action Distance (ICML 2026)
· Steccanella, L., Evans, J.B., Şimşek, Ö., and Jonsson, A. (2026). "Learning the Minimum Action Distance". In Proceedings of the 43rd International Conference on Machine Learning, PMLR.
We propose a novel method for learning the minimum action distance in reinforcement learning. Our approach uses quasimetrics to model asymmetric distances in directed environments, leading to more accurate estimates that improve performance on downstream tasks.
Creating Multi-Level Skill Hierarchies in Reinforcement Learning (NeurIPS 2023)
· Evans, J.B. and Şimşek, Ö. (2023). "Creating Multi-Level Skill Hierarchies in Reinforcement Learning". Advances in Neural Information Processing Systems, 36, pp.48472-48484.
We propose a novel characterisation of a multi-level skill hierarchy for reinforcement learning. Our approach uses modularity maximisation to expose the structure of the agent's interaction with its environment at multiple levels of abstraction.
Workshop Papers
Structural Analogies as Posterior Transfer in Continual Reinforcement Learning (GenPlan 2026)
· Traverso, S., Evans, J.B., Sprekeler, H., and Toussaint, M. (2026). "Structural analogies as posterior transfer in continual reinforcement learning: When does analogy help?". 9th Workshop on Generalization in Planning @ ICAPS.
We propose a novel method for transfer in continual reinforcement learning via graph matching.
Preprints
No papers currently available only as preprints.