Daniel Beechey
PhD Student
Department of Computer Science, University of Bath, United Kingdom
djeb20 AT bath DOT ac DOT uk
I am a PhD student at the Bath Reinforcement Learning Laboratory, University of Bath, supervised by Özgür Şimşek. I am also a member of the Centre for Doctoral Training in Accountable, Responsible and Transparent AI (ART-AI). My research interests lie in artificial intelligence, with a focus on reinforcement learning.
Research Summary: I am interested in the capacity of AI systems with limited resources to acquire generalisable behaviours that facilitate continual learning. Specifically, I focus on two main areas: discovering broad and generalised skills that allow for continual adaption to change and explaining the behaviour of these agents by revealing how they are influenced by their observations. For further details, please refer to the research section below.
News
- Sep. 2024: Co-organising the Bath RL Workshop.
- Sep. 2024: Attending the Mediterranean Machine Learning Summer School.
- Jul. 2024: Attending the CIFAR Deep Learning + Reinforcement Learning (DLRL) Summer School.
- Jun. 2024: Accepting a Doctoral Recognition Award at the Bath Doctoral Research Celebration Evening!
- Jun. 2024: Giving a talk as part of the Digital Futures session at the Bath Doctoral Festival of Ideas.
- Apr. 2024: Gave a talk to the Bath AI Society on the work of the Bath Reinforcement Learning Lab.
- Sep. 2023: Our work on "Reformulating Reactivity Design for Data-Efficient Machine Learning" has been accepted to ACS Catalysis!
- Aug. 2023: Presented our poster on "Explaining Reinforcement Learning with Shapley Values" at ICML 2023.
- Jul. 2023: Presented a talk on "Explaining Reinforcement Learning with Shapley Values" at the Bath Conference of Computer Science.
- Jun. 2023: Presented a talk on "Explaining Reinforcement Learning with Shapley Values" at the Alan Turing Institute Student Presentations.
- Apr. 2023: Our paper on "Explaining Reinforcement Learning with Shapley Values" has been accepted as a poster at ICML 2023!
- Oct. 2021: Joined the Bath Reinforcement Learning Lab at the University of Bath as a 1st year Ph.D. student.
Research
My research aims to (1) uncover general skills for agents operating under contrained resources, enabling continual adaptation to new and diverse tasks, and (2) explain the behaviour of agents in terms of policy and performance.
Explaining Agent Behaviour
An agent's behaviour can be explained by their policy, but this mapping from states to distributions over actions can be arbitraily opaque. How can we explain the behaviour of agents by revealing how they were influenced by the features of their observations?
Explaining Reinforcement Learning with Shapley Values [ICML 2023]Discovering General Skills
All forms of intelligence are constrained by time and computational resources. Humans minimise planning efforts by learning hierarchical behaviours, enabling quick adaptation to new challenges. How can an agent discover and learn a collection of skills to continually adapt to change?
Publications
Talks
An Introduction to Explainable and Hierarchical Reinforcement Learning
Bath AI Society, April 2024
Teaching
