While artificial intelligence is making ever-larger strides, we cannot neglect the potential for its impact on the way we nurture human intelligence. Education is without a doubt a primary candidate for a beneficial application of generative AI and machine learning. In this talk, Dr. Elitza Maneva will specifically concentrate on their application to the personalization of education.
Knowledge tracing is the process of modeling the knowledge state of learners. Knowledge space theory provides us with a graphical model for the learning process, so that this process, including the aspect of forgetting, can be represented as a random walk on a graph. We'll define a synthetic model of a student and evaluate how the teacher's knowledge of the structure of the knowledge space can affect the efficiency of the learning process. Student data and language models give us a way to discover the structure of this space in great detail. We propose applications of this model specifically to the areas of reading comprehension and programming literacy.
Elitza Maneva received her PhD in Computer Science from the University of California at Berkeley. Her doctoral thesis was on Belief-Propagation algorithms for Constraint Satisfaction Problems. She has published research in a wide variety of areas, including game theory, sensor networks, quantum and classical information theory, and computational complexity. She has carried out postdoctoral stays in IBM Almaden Research Labs, Universitat Politècnica de Catalunya, and Universitat de Barcelona.
In the last 8 years, Elitza has worked in educational game development and middle years math education. Her games are available on the AppStore under the brand of Balconia Math. She also recently completed a Masters degree in Secondary School Education. At UAB Elitza will work on applications of artificial intelligence and gamification to education and health.