Liang-Kang Huang

About

Liang-Kang Huang is a researcher at the intersection of robotics, natural language processing, and human-robot interaction. His primary contributions focus on enabling robots to learn complex tasks from human guidance, particularly through the integration of demonstrations and natural language instructions. His most cited work, "Reward Learning from Narrated Demonstrations" (2018, 6 citations), introduces a novel framework that allows robots to infer reward functions by combining physical demonstrations with accompanying verbal narrations. This approach moves beyond traditional methods that rely solely on visual goal states or pose specifications, capturing the rich, communicative way humans naturally teach one another. By leveraging language as a signal for intent and reward, Huang's work addresses a fundamental challenge in imitation learning: how to extract not just *what* a robot should do, but *why* it is doing it. This research has implications for making robotic programming more intuitive and accessible, bridging the gap between human communication and machine learning. His contributions are particularly notable for advancing the paradigm of learning from human interaction, where language serves as a powerful, efficient channel for transferring task knowledge.

Research Focus

Key Achievements

Top Papers

1. 1
   Reward Learning from Narrated Demonstrations

   6 citations · 2018

Key Collaborators

Contact & Links