Luke Dickens
@lukedickens.bsky.social
Associate Professor at UCL. Teaching on Knowledge Information and Data Science (KIDS) MSc. Member of KIDS (UCL) and SPIKE (Imperial/UCL) research groups.
Interested in a Masters degree in Knowledge, Information and data science? Come and join us for our KIDS MSc Open Day on 20th Nov. ‘25. More information and a chance to register here: www.ucl.ac.uk/arts-humanit...
Knowledge, Information and Data Science MSc: Virtual Open Event
Join our Graduate Open Event for anyone who is interested in applying for the Knowledge, Information and Data Science MSc at UCL.
www.ucl.ac.uk
November 18, 2025 at 2:48 PM
Interested in a Masters degree in Knowledge, Information and data science? Come and join us for our KIDS MSc Open Day on 20th Nov. ‘25. More information and a chance to register here: www.ucl.ac.uk/arts-humanit...
How might you swap materials or steps in a recipe, manufacturing process or other making procedure to achieve similar ends? We developed a formal, graph-based framework to represent and reason about such substitutions, and demonstrate it in ASP. See our JAIR paper: www.jair.org/index.php/ja...
A Graphical Formalism for Reasoning about Substitution in Resource Transforming Procedures
| Journal of Artificial Intelligence Research
www.jair.org
November 13, 2025 at 2:23 PM
How might you swap materials or steps in a recipe, manufacturing process or other making procedure to achieve similar ends? We developed a formal, graph-based framework to represent and reason about such substitutions, and demonstrate it in ASP. See our JAIR paper: www.jair.org/index.php/ja...
Three new PhD studentship opportunities in AI and machine learning at the KIDS research group (UCL, UK) with me and colleagues Antonis Bikakis and Vasileios Routsis. Closing date 5th Jan ‘26. Apply now. Details here: tinyurl.com/4x2uhxax
Department: Information Studies
tinyurl.com
November 13, 2025 at 2:08 PM
Three new PhD studentship opportunities in AI and machine learning at the KIDS research group (UCL, UK) with me and colleagues Antonis Bikakis and Vasileios Routsis. Closing date 5th Jan ‘26. Apply now. Details here: tinyurl.com/4x2uhxax
My department (DIS) at UCL (London, UK) is recruiting a full-time Lecturer with research profile in: AI, knowledge representation, machine learning, language technologies, and data science. Closing date: 24/6/25. Post starts: Oct'25. More details at: tinyurl.com/4zn5n5y5
UCL – University College London
UCL is consistently ranked as one of the top ten universities in the world (QS World University Rankings 2010-2022) and is No.2 in the UK for research power (Research Excellence Framework 2021).
tinyurl.com
June 13, 2025 at 11:31 AM
My department (DIS) at UCL (London, UK) is recruiting a full-time Lecturer with research profile in: AI, knowledge representation, machine learning, language technologies, and data science. Closing date: 24/6/25. Post starts: Oct'25. More details at: tinyurl.com/4zn5n5y5
My colleague Kexin Baugh is presenting our paper: Neural DNF-MT: A Neuro-symbolic Approach for Learning Interpretable and Editable Policies at AAMAS this week. Access the preprint here: arxiv.org/abs/2501.03888
Neural DNF-MT: A Neuro-symbolic Approach for Learning Interpretable and Editable Policies
Although deep reinforcement learning has been shown to be effective, the model's black-box nature presents barriers to direct policy interpretation. To address this problem, we propose a neuro-symboli...
arxiv.org
May 20, 2025 at 3:24 PM
My colleague Kexin Baugh is presenting our paper: Neural DNF-MT: A Neuro-symbolic Approach for Learning Interpretable and Editable Policies at AAMAS this week. Access the preprint here: arxiv.org/abs/2501.03888
Behavioural alignment is an alternative to representational alignment to measure information processing similarities, and can be used more readily with human studies and data. We present new metrics and evaluate against existing ones: arxiv.org/abs/2409.13919
Measuring Error Alignment for Decision-Making Systems
Given that AI systems are set to play a pivotal role in future decision-making processes, their trustworthiness and reliability are of critical concern. Due to their scale and complexity, modern AI sy...
arxiv.org
January 29, 2025 at 3:04 PM
Behavioural alignment is an alternative to representational alignment to measure information processing similarities, and can be used more readily with human studies and data. We present new metrics and evaluate against existing ones: arxiv.org/abs/2409.13919
Hi, I'm Luke Dickens and I'm new to Bluesky! I work as an Associate Professor in Machine Learning and AI at University College London (UCL). I plan to discuss my research interests including topics in Trustworthy AI, and to promote the KIDS and SPIKE research groups of which I am a member.
January 29, 2025 at 2:35 PM
Hi, I'm Luke Dickens and I'm new to Bluesky! I work as an Associate Professor in Machine Learning and AI at University College London (UCL). I plan to discuss my research interests including topics in Trustworthy AI, and to promote the KIDS and SPIKE research groups of which I am a member.