You investigate how agents learn through interaction, simulation, and structured feedback, represent also manipulate knowledge in compositional forms, as well as integrate reinforcement learning with symbolic abstractions, hierarchical planning, memory, and reasoning. + strong expertise in reinforcement learning and agentic AI, including sequential decision‑making and learning‑based planning + deep understanding of goal‑directed AI systems involving memory, tool use, planning, multi‑step reasoning, and long‑horizon behavior + ability to design systems that organize knowledge in semantically meaningful ways while supporting action, planning, interpretability, and generalization + ability to frame complex technical challenges in terms of sequential decision‑making, planning, or knowledge‑based reasoning
mehr