Uljad Berdica

I am a 2022 Rhodes Scholar and PhD student in the CDT-AIMS program at the University of Oxford focusing on Reinforcement Learning, World Models, Exploration and Agentic LLMs. Working on meaningful AI that learns from experience, under the supervision of Prof. Jakob Foerster and Prof. Perla Maiolino from the Oxford Robotics Institute.

SR at Google DeepMind working on Game Theory, Reinforcement Learning LLMs.

My most recent papers include methods to make LLMs more effective and diverse, unification of Offline RL Algorithms to automatically discover new ones and bioimpedance measurement in Electronics. Currently interested in Backprop-Free methods and automating research with AI. Previously interned at J.P. Morgan’s AI Research group.

Served as a reviewer for IEEE Robotics Journals, conferences like ICML, ICLR, AAAI, IEEE RoboSoft and numerous workshops. Since moving out of my home country at 17 on a scholarship, I have lived in the USA, UAE, China, and France as part of my studies before moving to the UK for my PhD. I love doing stand-up comedy.