Our Researchers

Meet the researchers

Current doctoral researchers

The MINDS Centre for Doctoral Training is no longer recruiting. Its doctoral researchers are completing their research, and this site shares their work and the programme’s wider activity.

Epifanios Baikas

Developing privacy-preserving AI that can learn directly on memory-constrained Internet of Things devices, reducing reliance on cloud processing.
Tyler Clark

Tyler Clark is demonstrating that high-performance deep reinforcement learning can run on standard desktop computing, reducing reliance on expensive large-scale infrastructure.
Xiyue Fan

Training transformer models on mathematical expressions to investigate whether AI can approximate solutions to differential equations in a symbolic form that may be easier to inspect than numerical output alone.
Oliver Grainge

Oliver Grainge is developing compact visual AI systems that help robots and drones recognise where they are when GPS is unavailable, using hardware with limited memory and computing power.
Balint Gucsi

Developing social robots that adapt their decisions and behaviour to individual users, with the aim of making human-robot interactions more useful and less frustrating.
Jack Hilton-Jones

Jack Hilton-Jones is developing predictive-control methods for flexible magnetic robots, aiming to help them navigate blood vessels more safely and precisely during minimally invasive procedures.
Sam Hilton-Jones

Using mathematical analysis to investigate how transformer models focus and reason across long sequences, helping to open the black box of modern large language models.
William Hunt

Developing interfaces and language-based control methods that help a single human operator understand, communicate with and oversee swarms of autonomous robots.
Xander Lowe

Combining laboratory electrophysiology and computational brain modelling to investigate short-term memory, with potential relevance to neurological conditions and low-energy, brain-inspired computing.
Kieran Maguire

Investigating how patterns across networks shape the link-prediction models behind many online recommendations, with the aim of making recommendations more relevant, transparent and fair.
Rudra Mutalik

Helping large language models learn new information without forgetting existing knowledge, with the aim of keeping AI systems up to date without extensive, costly retraining.
Yaseen M. Osman

Investigating how large language models can run more efficiently on lower-compute devices, with potential benefits for privacy, sustainability and wider access to AI.
Keniel Peart

Developing AI that interprets sign-language movement from video and radar using 3D human modelling, with potential applications in assistive communication and privacy-aware sensing.
Michail Rontionov

Michail Rontionov is developing compiler tools that translate neuromorphic algorithms into digital FPGA hardware, aiming to bridge the gap between brain-inspired computing research and physical devices.
Lianghong Ye

Lianghong Ye is investigating whether tiny mechanical resonators can process time-dependent data, with the aim of supporting lower-power physical computing for sensors and edge AI.
Nehal Yasin

Investigating shortcut learning, where machine learning models rely on the easiest patterns in their data, and how models could be helped to learn more diverse and robust patterns.

MINDS alumni

These researchers have completed their doctoral research with MINDS. Their profiles describe that research and, where they have shared it, what they do now.

Agent-Based Adaptive Systems

Decentralised AI for Situational Awareness

Thomas Kelly
Game Theory and Multi-Agent Systems to Detect and Respond to Cyber Intrusions in IoT Environments

Charles Hutchins
Reducing Carbon Emissions in Shipping: Tax Incentives, Optimisation and Prediction

Matthew Durrant
Emergent Communication in Social Games

Olaf Lipinski
A Neuro-Symbolic Approach to Explainable Reinforcement Learning

Mark Towers
Nature-Inspired Communication in Reinforcement Learning Systems

Jennifer Barnes-Nunn
Efficient agent training from Human Preferences and Justifications in safety-critical environments

Ilias Kazantzidis
Interactive Lifelong Reinforcement Learning (Cohort A)

John Birkbeck
Interactive Lifelong Reinforcement Learning (Cohort B)

Kryspin Varys
Radar Imaging for Non-Invasive for Brain Imaging and Anamoly Detection

Keniel Peart
Intelligent Coordination and Planning for Human-Robot Swarms

William Hunt
Sample-Efficient Exploration in Visually Complex Reinforcement Learning Tasks

Tyler Clark
Multi-modal event classification of social media content where training data is limited and/or emerging

Rudra Mutalik

Embedded Artificial Intelligence

Achieving Progressive Intelligence for Low Power Systems

Jack Dymond
Relative Movement Quantisation for Lossless Network Compression

Christopher Subia-Waud
Binaural Machine Listening in the Cocktail Party Scenario

Hsuan-Yang Wang
Graph Neural Networks for Robotic Grasping

Kian Spencer
Convolutional Neural Networks for Auditory Attention in Robots

Elliot Stein
Efficient learning on resource-constrained embedded systems

Epifanios Baikas
Machine learning for the analysis of ambient marine noise

Konstantinos Drylerakis
Federated Learning on Embedded Systems

Shannon How
Modelling Quantum Chromodynamics with Geometric Deep Learning

Kieran Maguire
Secure and Resilient Hardware Implementation of AI Modules

Arezou Nayebi
REACTOR: Robust Embedded Adaptive Control Techniques for Online Reconfiguration

Carl Richardson
Disentangled Representation Learning with Applications in Marine Environments

Nikos Chazaridis
Creating Augmented Reality Soundscape Experiences on Edge Devices

Mark Saunders
Efficient Visual Place Recognition for Real-Time Visual Perception

Oliver Grainge