PhD candidate in Computer Science at the University of Leicester, researching graph machine learning, Deep Graph (graph-to-graph) Transformation, benchmarking, and reproducible model evaluation. Thesis submission expected September 2026.
Featured research — edge-dgt-eval
Open-source architecture evaluation framework for graph-to-graph learning, developed as a key contribution of my PhD research.
It enables controlled comparison of non-parametric, feedforward, convolutional, attention-based, and graph-native architectures under shared experimental conditions, across synthetic and real-world graph learning tasks.
Research focus: graph transformation · benchmarking · reproducibility · architecture comparison · relational inductive bias
- Benchmark First: Defining Tasks for Graph Transformation Learning — ICGT 2026
- Graph Rewriting for Graph Neural Networks — ICGT 2023 · Best Paper Award nominee