AI-driven personalised support to learning beyond problem solving

There is increasing interest in AI-based environments to support educational activities, such as those more exploratory in nature, like learning from interactive simulations or playing educational games. However, not all learners benefit may benefit from AI-driven personalised support that can help those who need it while maintaining the unconstrained nature of the interaction.

In this talk, Cristina will discuss the unique challenges of this endeavour and a proposed solution based on data-driven AI models that can detect and react to behaviors indicating that a learner is struggling with the exploratory process. This will include a summary of results on applying this approach to various exploratory environments and on making such models explainable to their end users, with explanations personalised to each user’s specific needs.

This event will be particularly useful for researchers, educators, teachers and students.

Please note this is a hybrid event and can be joined either in-person or online.

Related links

• UCL Knowledge Lab
• Culture, Communication and Media