← all projects
ARIA Opportunity Space: Nature Computes Better
AI for Adaptive Curricula for Human Learning
Humans can rapidly discover useful strategies in unfamiliar environments, but the computational mechanisms that drive this "discovery" remain poorly constrained: very different models can fit the same behavioral data. This project measures how people explore and learn in a set of controlled, video-game-like environments, and uses these data to build adaptive, individualized curricula that accelerate learning—by selecting the next tasks, environment settings, or information that best supports each learner's progress. To resolve ambiguity between competing theories of exploration (e.g., goal-directed planning, intrinsic motivation, heuristic search, or uncertainty-driven sampling), we record neural signals during exploration and use them to ground model inference. By linking brain dynamics to computational models we aim to identify which mechanisms best explain human discovery, and leverage those insights to design curricula that are both effective and interpretable.
Rui Ponte Costa's Lab
University of Oxford
,
Oxford
Costa Lab
↗
Csabi holds a PhD in core ML (focusing on Continual and Self Supervised Learning) from University of Oxford with collaborations at Apple, Intel and Meta. Prior to his fellowship he worked on Multi-Modal World Models as an ML engineer for Silent Creek, a High Frequency Trading firm.
Rui Ponte Costa
Group Leader, Centre for Neural Circuits and Behaviour, Department of Physiology, Anatomy and Genetics
Rui Ponte Costa leads the Neural & Machine Learning Group at Oxford, bringing together neuroscience, psychology and machine learning.
Chris Summerfield
Professor of Cognitive Neuroscience at the University of Oxford, Research Director at the UK AI Security Institute
Chris' work focuses understanding the cognitive and neural mechanisms that underlie human learning and decision-making, and on studying the impacts of AI on society. His research bridges the fields of cognitive science, neuroscience, and artificial intelligence. He is particularly interested in how insights from human cognition can inform the development of more advanced and safer AI systems. Chris also lead the Human Information Processing (HIP) lab in the Department of Experimental Psychology at the University of Oxford.
