Yoonsuck Choe

Texas A&M University, The University of Texas at Austin

Papers

Total Citations

H-Index

About

Yoonsuck Choe is a leading researcher in developmental robotics, computational neuroscience, and autonomous mental development. His work explores how agents—biological or artificial—can learn from scratch through sensorimotor interaction, without pre-programmed knowledge. Choe’s major contributions center on understanding the earliest stages of learning: how sensory and motor primitives form the foundation for higher cognition. His 2007 paper, “Autonomous Learning of the Semantics of Internal Sensory States Based on Motor Exploration” (40 citations), is a seminal work that argues developmental programs must first learn the meaning of their own internal states through active exploration, not external labels. This framework has influenced theories of embodied cognition and autonomous AI. In his 2012 study on binocular depth estimation (9 citations), Choe demonstrated how a humanoid robot can improve visual perception through simple, infant-like actions—showing that interaction, not just passive data, drives learning. His earlier modeling of self-organization in the visual cortex (1999, 4 citations) laid groundwork for understanding how neural maps emerge from experience. Choe’s work bridges robotics and neuroscience, offering a principled path toward truly autonomous, self-learning systems.