Julio H. Zaragoza

About

Julio H. Zaragoza is a researcher in robotics and artificial intelligence, with a primary focus on reinforcement learning for real-world service robots. His work addresses critical challenges in applying reinforcement learning to physical systems, particularly the difficulty of handling high-dimensional sensor data and the need for continuous, rather than discrete, action spaces. Zaragoza’s major contribution is the development of TS-RRLCA (Two-Stage Relational Reinforcement Learning with Continuous Actions), a novel framework that combines behavioural cloning with locally weighted regression. This two-stage method enables robots to learn complex tasks more efficiently by first imitating demonstrations and then refining policies through continuous-action reinforcement learning. His most cited paper, "Relational Reinforcement Learning with Continuous Actions by Combining Behavioural Cloning and Locally Weighted Regression" (2010, 6 citations), and its precursor from 2009, lay the groundwork for more practical, data-efficient robot learning. While his citation counts are modest, Zaragoza’s work is notable for bridging the gap between theoretical reinforcement learning algorithms and the constraints of real robotic systems, offering a pathway toward more adaptive and capable service robots.

Research Focus

Key Achievements

Top Papers

1. 1
   Relational Reinforcement Learning with Continuous Actions by Combining Behavioural Cloning and Locally Weighted Regression

   6 citations · 2010
2. 2
   A Two-Stage Relational Reinforcement Learning with Continuous Actions for Real Service Robots

   2 citations · 2009

Key Collaborators

Contact & Links