Joshua B. Evans

Joshua B. Evans is a Lecturer in the Department of Computer Science at the University of Bath, Faculty of Science, where he is an active researcher and educator in artificial intelligence and reinforcement learning. He is affiliated with the Bath Reinforcement Learning Laboratory and serves as unit convenor for key on-campus and online reinforcement learning courses (CM50270, CM30359, and CM500336).

His research centers on enabling AI agents to learn complex behaviors through interaction, particularly via skill discovery in reinforcement learning. Key interests include:

• Hierarchical Reinforcement Learning
• Graph-Based Skill Discovery Methods
• Nature and Human-Inspired Skill Discovery
• Intrinsically-Motivated Reinforcement Learning
• World Models and Representation Learning

His recent publications, including a NeurIPS 2023 paper on multi-level skill hierarchies, emphasize graph-based and modular approaches to structuring agent behavior across abstraction levels. The work demonstrates strong theoretical grounding and practical application in diverse environments.

Scientific recognition includes:

• Doctoral Recognition Award, Bath Doctoral Research Celebration Evening
• Bill Tutte Scholarship from the Bill Tutte Memorial Fund

Dr. Evans actively supervises undergraduate and postgraduate students on a wide range of reinforcement learning projects, from game AI to healthcare applications and multi-agent systems. He has developed and delivered both on-campus and online MSc AI curricula, significantly contributing to AI education at Bath. His research includes software contributions such as SimpleOptions, SimpleEnvs, and OfficeWorld, supporting reproducibility and experimentation in HRL.

He is based at 1W 4.58, Department of Computer Science, University of Bath, BA2 7AY, UK.