Tom Jurgenson

About

Tom Jurgenson is a researcher working at the intersection of reinforcement learning, robotics, and motion planning, with a focus on developing principled frameworks for goal-directed AI systems. His most notable contribution, the Sub-Goal Trees framework, introduces a novel approach to representing and optimizing trajectories in goal-based settings, bridging reinforcement learning, imitation learning, and trajectory prediction. This work, which appeared in multiple forms across 2019 and 2020, addresses a fundamental challenge in AI: enabling agents to efficiently reason about complex, multi-stage goals rather than singular objectives. Jurgenson has also made meaningful contributions to neural motion planning, exploring how reinforcement learning can be harnessed to accelerate solutions to robotic motion problems by leveraging previously solved instances. His research is particularly relevant to real-world robotics applications, where efficient and generalizable planning is critical. With citations accumulating across his early career publications, Jurgenson's work is gaining recognition within the reinforcement learning and robotics communities. His research offers valuable tools for students and practitioners seeking to build more flexible, goal-aware autonomous systems capable of operating in complex environments.

Research Focus

Key Achievements

Top Papers

1. 1
   Sub-Goal Trees -- a Framework for Goal-Based Reinforcement Learning

   8 citations · 2020
2. 2
   Harnessing Reinforcement Learning for Neural Motion Planning

   6 citations · 2019
3. 3
   Sub-Goal Trees -- a Framework for Goal-Directed Trajectory Prediction and Optimization

   6 citations · 2019

Key Collaborators

Contact & Links