- Joanna Bryson (personal page)
- Marina De Vos (personal page)
- Kwang In Kim
- Julian Padget (personal page)
- Mike Tipping (personal page)
- Nataliya Mogles (personal page)
- Javier De La Dehesa Cueto-felgueroso
- Kewei Duan
- Swen Gaudl (personal page)
- Charlie Page
- Thomas Smith
- Andreas Theodorou (personal page)
- Matthew Thompson
- Siriphan Wichaidit
- Robert Wortham (personal page)
- Esraa Alwan
- Vincent Baines
- Tina Balke (external staff)
- Gideon Dadik Bibu
- Ajith Raam Dhoraisawamy
- Kewei Duan
- Joao Duro
- Emad Eldeen Elakehal
- Bidan Huang
- Fatemeh Jahedpari
- Jason Leake (personal page)
- Jeehang Lee (personal page)
- Tingting Li (personal page)
- Gokhan Mevlevioglu
- Jekaterina Novikova (personal page)
- Tim Pinder
- Saeid Pourroostaei Ardakani
- Paul Rauwolf (personal page)
- Zoreh Shams
- Daniel Taylor (personal page)
- Dimitrios Traskas
- Yifei Wang (personal page)
- Andrew Watson
Our group explores the two-way relationship between natural and artificial intelligence, with an emphasis on constructing and modelling complete systems.
Intelligence is all around us, but not everywhere around us. In both nature and artefact, we find a fantastic diversity of intelligence, starting from direct connections of simple sensing to simple action like we see in thermostats setting heating or plants growing towards the sun, and extending to complex systems of humans, our culture and our intelligent machines.
Today intelligent systems (often using machine learning) recognise our voice commands, correct our spelling, suggest music and films we like, and search for stories we remember from just a few key words. They help us drive our cars and fly our planes. They learn to predict our desires and behaviours, and can even help us do science or art.
None of the achievements of artificial intelligence come from a single algorithm, equation or data set. Rather, developing machines with intelligent behaviour requires constructing a system of interacting parts: learning, sensing, memory, action, interaction, planning, reasoning, users, professionals.
The Intelligent Systems Group is the newest group in the Department of Computer Science with some very exciting recent hires.
There are usually opportunities for postgraduate research in our group. If you need funding, the best time to apply is by mid-December of the year before you want to start. Funded studentships normally start in the October. However, sometimes funding comes available on short notice. To apply, first contact members of staff who share your interests, or look at their proposed PhD projects posted here.
Current research activities for this group can be placed under four broad headings.
Agency and Societies
- The Energy Project: sensors to build models of occupant behaviour to improve sustainability.
- Understanding bond formation between members of groups, how an individual's identity and investment can be spread between groups, and how in-group / out-group dynamics affect investment in social goods.
- Culture, communication, gossip & deception.
- Personalities and social roles, and their influence on social cognition and group behaviour.
- The evolution of social structures, and of cognition more generally.
- Political polarisation and the causal explanation of its correlation with income inequality.
Autonomous Robots and Game AI
- Real-time grasping and dexterous manipulation.
- Synthetic emotions, and their impact on Human Robot Interaction.
- Coordination and mapping with AUVs: path planning driven by map and 3D image construction.
- Robot and AI ethics: the ethical design of intelligent systems and their role in society.
- Intelligent and adaptive non-player characters (NPCs) for games.
- Intelligent assistance for level design, and procedural generation of game content more generally.
- Learning on data manifolds: semi-supervised learning, spectral clustering, non-linear data embedding, link prediction.
- Bayesian inference: Large-scale approximate Bayesian inference, latent variable models.
- Learning for computer graphics: Bayesian inference for shape modelling, tracking, sampling, and transfer, machine learning for computational photography, videography, and 3D data analysis.
- Sparse Bayesian models (the “relevance vector machine”) and related novel learning techniques.
- Probabilistic approaches to tree-based pattern recognition.
- Adaptive analysis of multivariate time series.
- Methods for intelligent statistical automation.
- New perspectives on deep neural networks.
- Model-driven data mapping and visualisation techniques.
Reasoning and Representing
- Music composition and performance.
- Legal reasoning.
- Task learning & systems neuroscience.
- Answer Set Programming (ASP).
- Norms and institutions as mechanisms for analysis and control.
- Sensor networks.
Seminar series and mailing lists
We have an active seminar series with local, national and occassionally international speakers. There are also many relevant seminars and collaborations around campus.
To be kept up to date (whether you are from the University of Bath or not), add yourself to any of the following mailing lists on AI, natural intelligence, robots, mathematical biology, networks and collective behaviour, or evolutionary social sciences talks at or around the University of Bath:
- bai (artificial intelligence)
- amoni (modelling natural intelligence)
- robotics, cmb (mathematical biology)
- cncb (networks and collective behaviour)
- bunn (neuroscience)
- bess (evolutionary social science)
- events at the Institute for Mathematical Innovation