Dr Paul Duckworth is a post doctoral research assistant at the Oxford Robotics Institute. He is working in the GOALS group with Prof Nick Hawes combining probabilistic machine learning with goal-oriented planning for improved autonomous mobile robot behaviour.
Prior to joining the ORI he was a PDRA in the Machine Learning Research Group with Professor Mike Osborne. He received his PhD (2017) from the University of Leeds, working with Prof Anthony G. Cohn and Prof David C. Hogg, on machine learning for autonomous mobile robots as part of the STRANDS project. He was awarded an MSc in Mathematics and Computational Science from the University of Manchester (2010), and a BSc in Mathematics and Statistics from Lancaster University (2008).
His current research interests lie in probabilistic machine learning and mobile robotics, and he is also interested in the impact machine learning technologies have on society and the future of human work, and our understanding of algorithmic creative intelligence.
In a previous life he was a statistical programmer for a pharma CRO, and also a banking analyst.
Google Scholar: https://scholar.google.co.uk/citations?user=tQzXW4QAAAAJ&hl=en
Awards and Achievements
- 2017 IJCAI Best Video Award “Grounding of Human Environments and Activities for Autonomous Robots”, Duckworth, P., Alomari, M., Hogg, D.C., Cohn, A.G., In International Joint Conference on AI (IJCAI) 2017.
- 2017 RoboNLP Best Paper Award “Natural Language Grounding and Grammar Induction for Robotic Manipulation Commands” Alomari, M., Duckworth, P., Hogg, D.C., Cohn, A.G., In ROBONLP Workshop at ACL 2017.
- Best Paper. 2016 ECAI Best Student Paper Award - Runner-up. “Unsupervised Activity Recognition using Latent Semantic Analysis on a Mobile Robot” Duckworth, P., Alomari, M., Gatsoulis, Y., Hogg, D.C., Cohn, A.G., In European Conference on AI 2016.
Paul's research interests are broad and split between probabilistic machine learning for mobile robotics, and AI for social good themes such as the future of AI and human work.
- "Markov Decision Processes with Unknown State Feature Values for Safe Exploration using Gaussian Processes", Budd, M., Lacerda, B., Duckworth, P., and Hawes, Nick, in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020.
- "Unsupervised human activity analysis for intelligent mobile robots", Duckworth P., Hogg, D.C., Cohn, A.G., In Artificial Intelligence 270, 67-92, 2019.
- “Inferring Work Task Automatability from AI Expert Evidence”, Duckworth, P., Graham, L., and Osborne, M. In AI Ethics & Society AAAI/ACM 2019. Also appears in: AI for Social Good Workshop at NeurIPS 2018.
- "Qualitative and Quantitative Approach to Assess of the Potential for Automating Administrative Tasks in General Practice" Willis, M., Duckworth, P., Coulter, A., Meyer, E., Osborne, M. in BMJ Open 10 (6) 2020.
- “The Future of Healthcare Protocol Article” Willis, M., Duckworth, P., Coulter, A., Meyer, E., Osborne, M. In Journal of Medical Internet Research (JMIR) Protocols 2019.
- "Latent Dirichlet Allocation for Unsupervised Activity Analysis on an Autonomous Mobile Robot", Duckworth, P., Alomari, M., Charles, J., Hogg, D.C., Cohn, A.G. Thirty-First AAAI Conference on Artificial Intelligence 2017