The next frontier in terms of generative ai is the creation of coherent videos from text. Video is a type of data that exists in many shapes and forms, and as such the generation of videos is a very large problem which I study.
I am interested in creating algorithms that allow to map brain activity to actual images/videos seen by the user. Usually, this is done via fMRI.
I am interested in modifying generative algorithms such as Stable-Diffusion to produce interesting aesthetics.
Previously, my research focused on autonomous agents and Reinforcement Learning.