Derek Guo

Papers

Total Citations

H-Index

About

Derek Guo is a rising researcher at the intersection of robotics and machine learning, with a primary focus on enabling generalizable robotic reinforcement learning (RL) through large-scale, data-driven methods. His most notable contribution is the development of value-function learning techniques that allow robotic systems to pre-train on vast, heterogeneous datasets—including internet videos—and then fine-tune for specific tasks. This work, detailed in his highly cited 2024 paper "Robotic Offline RL from Internet Videos via Value-Function Learning," tackles a fundamental challenge: how to imbue robots with the broad generalization capabilities seen in modern ML systems, such as large language models. By demonstrating that offline RL can effectively leverage prior experience from diverse sources, Guo's research paves the way for robots that learn more efficiently and adapt to novel environments without requiring extensive, task-specific data collection. His approach has garnered significant attention (5 citations in its first year) and is considered a key step toward scalable, real-world robotic learning. Guo's work is essential reading for anyone interested in the future of generalist robots and data-efficient RL.