Nicklas Hansen develops data-driven world models (models of the physical world) that enable robots to "understand" and interact with the real world. While we have seen rapid progress in AI over the last few years, teaching AI agents to reliably interact with our physical world has proven difficult and, consequently, this new paradigm of large models trained on large datasets has not materialized as much in robotics as in related areas (natural language and computer vision) that operate in our digital lives. His overarching research goal is to build generalist robots that autonomously perform tasks despite the messiness of our real world, generalize to new scenarios that they have not encountered before, and -- importantly -- learn from those new experiences.
Nicklas Hansen is a third-year computer science PhD student at the University of California San Diego advised by Hao Su and Xiaolong Wang. He is interested in building generally capable AI systems that operate in real world environments. Prior to his PhD, he obtained BS and MS degrees in computer science from Technical University of Denmark, and also spent time at the University of California Berkeley.