We are delighted to announce that Dr Retno Larasati successfully defended her thesis today.
Retno’s study was entitled: Trust and Explanation in Artificial Intelligence Systems: A Healthcare Application in Disease Detection and Preliminary Diagnosis.
In her work, Retno has looked at the extent to which explanations and interactions can help non-expert users properly calibrate trust in AI systems, specifically AI for disease detection and preliminary diagnosis. This means reducing trust when users tend to over-trust an unreliable system and increasing trust if the system can be shown to work well. The three fundamental contributions to knowledge are, first, informing how to construct explanations that non-expert users can make sense of (meaningful explanations). Second, it contextualises current AI explanation research in healthcare, informing how explanations should be designed for AI-assisted disease detection and preliminary diagnosis systems (Explanation Design Guidelines). Third, it provides preliminary insights into the importance of the interaction modality of explanations in influencing trust. Her preliminary findings can inform and promote future research on XAI by shifting the focus of current research from explanation content design to explanation delivery and interaction design.
The examiners, Dr. Paul Mulholland and Dr. Simone Stumpf (University of Glasgow), said they were impressed with Retno’s work and her contribution to knowledge.
Her supervisors were Professor Anna De Liddo and Professor Enrico Motta. They said: “There is nothing more rewarding for a supervisor than to see a student grow in confidence and achieve what they have worked so hard for. Retno has demonstrated a level of resilience and strength that, no doubt, will allow her to make a great contribution to any research group in the future. Well done, Retno, we are very proud of you!”
Looking ahead, Retno will take a Post-Doc position at KMi working with Dr. Tracie Farrell as part of her UKRI Future Leader Fellowship to study how Artificial Intelligence (AI) can help create a more "just" society, specifically examining the impact on marginalized groups.
This will no doubt give Retno the opportunity to maintain close ties with the IDea (Intelligent Deliberation team) within which she developed her doctoral work and with which she intends to continue collaborating in the future.
We are very happy to have retained Retno’s brilliant mind for a while longer in our network of young talents at KMi!
Time for the Lab to celebrate another great success!