Papers
6
Total Citations
135
H-Index
4
About
Yilun Zhou is a researcher at the intersection of artificial intelligence, robotics, and neuromorphic computing, with a focus on enabling intelligent agents to learn, adapt, and collaborate in human environments. His most cited work, "Self-powered high-sensitivity sensory memory actuated by triboelectric sensory receptor for real-time neuromorphic computing" (99 citations), pioneers a novel approach to sensory processing that mimics biological systems, offering breakthroughs in energy-efficient, real-time AI hardware. Zhou has also made significant contributions to procedural knowledge acquisition, as seen in his work on learning household task knowledge from WikiHow descriptions (16 citations), which advances commonsense reasoning for domestic robots. His research on incorporating side-channel information into convolutional neural networks (8 citations) enhances robotic perception, while his development of adversarially guided self-play (6 and 4 citations) addresses the critical challenge of social convention adoption and latent space alignment in multi-agent systems. Additionally, his work on Robot Controller Understanding via Sampling (RoCUS) provides tools for interpreting robot behavior, bridging the gap between optimization and human comprehension. Zhou’s diverse portfolio—spanning hardware, knowledge representation, and multi-agent coordination—demonstrates a commitment to creating robots that are not only intelligent but also socially aware and transparent.
Research Focus
Key Achievements
Top Papers
- 1
- 2Learning Household Task Knowledge from WikiHow Descriptions16 citations · 2019
- 3
- 4Adversarially Guided Self-Play for Adopting Social Conventions6 citations · 2020
- 5Latent Space Alignment Using Adversarially Guided Self-Play4 citations · 2022
- 6RoCUS: Robot Controller Understanding via Sampling2 citations · 2020