GPTSwarm: evolutionary graph optimisation
Reproducing adversarial multi-agent LLM experiments and comparing REINFORCE with an evolutionary edge optimiser.
Team projectReport
University of Groningen · Academic work
Research projects
Studies in machine learning, reasoning and human-centred AI. Each project includes its methods, findings, limitations and full reports.
Browse the report librarySentenceBot
An Alpha Mini prototype combining tangible topic selection, adaptive difficulty and language-model feedback.
Team project2 reports
Fairness & explainability
Looking beyond prediction scores to understand how fairness interventions change decisions and feature attributions.
Team project2 reports
Conditioning diffusion transformers
A controlled comparison of adaptive layer normalisation and cross-attention for text-conditioned image generation.
Team projectReport
Dreaming agents
Tabular reinforcement learning with replay and perturbed experience in a small GridWorld.
Individual projectReport
Sign language recognition
Transfer learning with MobileNetV2, evaluated on people held out from training.
Team projectReport
Cognitive modelling with ACT-R
A declarative-memory model examining how social comparison changes contributions in a public goods game.
Individual projectReport
Recursive theory of mind
Bounded reasoning about what agents know, observe and believe about one another.
Individual projectReport
Multi-agent social simulation
Analysing relationships and survival dynamics in a constructed social environment.
Team projectReport
Responsible AI for urban traffic
A conceptual traffic-management proposal examining privacy, stakeholder needs and human oversight.
Team design studyReport