Explainable AI (XAI)
I am particularly interested in explainable reasoning using symbolic AI, working specifically on computational argumentation for making and explaining decisions using mined data, with applications to mathematical optimisation, clinical decision support, legal reasoning. XAI research currently spans across many of my activities.
ROAD2H
EPSRC project: Resource Optimisation, Argumentation, Decision support and Knowledge Transfer to create Value via Learning Health Systems. I apply computational argumentation methods to explainable reasoning with clinical guidelines and integration with optimisation.
Analysing the Passage of UK Parliament Bills
Project in collaboration with Thomson Reuters, Data Science Institute and Imperial College Business School. I have driven the development of both methodology and a working system to explain primary legislation outcomes.
Other activities
I have initiated Explainable AI seminar series at Imperial.
I engaged with UK policy-makers at
The Forum: AI & Health on March 3.
I have participated at an Explainability Expert Roundtable discussion on Explainable AI, at
The Alan Turing Institute.
I am a highly distinguished reviewer for two years in a row at
IJCAI (2018 & 2019), a premier AI conference.
I appear in Imperial's
Industry issue on Explainable AI.