Jinwook Oh

Korea Advanced Institute of Science and Technology

Papers

Total Citations

134

H-Index

About

Jinwook Oh is a leading researcher in energy-efficient computing and intelligent hardware design, with a focus on approximate computing and bio-inspired neural architectures. His seminal work, "Approximate computing: Challenges and opportunities" (2016, 101 citations), has become a foundational reference in the field, demonstrating how approximation techniques can dramatically reduce computational costs for data analytics and cognitive applications without sacrificing output quality. Oh’s contributions extend to pioneering hardware implementations for artificial intelligence, including a 57mW embedded mixed-mode neuro-fuzzy accelerator (2011) that enabled intelligent processing in power-constrained mobile and robotic platforms. He also developed a 201.4 GOPS real-time multi-object recognition processor (2009), which applied bio-inspired visual attention mechanisms to achieve high-performance, low-power object recognition—a significant leap from prior single-object systems. With over 130 citations across his most influential works, Oh’s research bridges the gap between algorithmic innovation and practical, energy-efficient silicon, making him a key figure in the advancement of intelligent, resource-aware computing systems for real-world applications.