Tzung-Feng Lin

Papers

Total Citations

H-Index

About

Tzung-Feng Lin is a researcher specializing in reinforcement learning and multi-agent systems, with a particular focus on advancing Q-learning for real-world robotic applications. His work addresses a critical limitation of conventional Q-learning: its reliance on pre-defined, discrete state and action spaces, which hinders performance in continuous, real-world environments. Lin’s major contributions include the development of Q-Learning with FCMAC (Fuzzy Cerebellar Model Articulation Controller) for multi-agent cooperation, which enables more adaptive and precise behavior in cooperative robotic tasks. His 2006 paper on this topic has garnered 7 citations, while his 2008 work on Q-learning in multi-agent cooperation and his 2012 paper on continuous action generation for Q-learning each have 5 citations. These studies collectively demonstrate his efforts to bridge the gap between theoretical reinforcement learning and practical robot control, allowing for smoother, more nuanced action selection. Lin’s research is particularly notable for its focus on real robot applications, where discrete action sets often fail to capture the subtle variations required for effective cooperation. His work remains relevant for researchers exploring continuous control and multi-agent coordination in robotics.