My long-term goal is to develop systems capable of open-ended scientific knowledge discovery. My current interests revolve around (in decreasing order of active involvement):

• Autonomous information seeking: uncertainty self-awareness, action-taking attending to information gain maximization, goal-directed exploration.
• Reinforcement learning for large models: efficient exploratinon capabilities in large search-spaces, open-ended & intrinsic objectives.
• Predictive world models: providing scalable streams of experiential data, grounding long reasoning trajectories and alleviating hallucinations.
• Knowledge representations: formal languages (e.g., math or code), hierarchical, and compositional representations.