Joonho Kong
Papers
1
Total Citations
8
H-Index
1
About
Joonho Kong is a leading researcher in computer architecture and systems, with a primary focus on the intersection of deep learning acceleration and hardware-software co-design. His work is instrumental in understanding how emerging neural network models perform on modern GPU platforms, particularly when subjected to quantization—a critical technique for reducing model size and inference latency. In his highly cited 2023 study, "Deep Learning Performance Characterization on GPUs for Various Quantization Frameworks," Kong systematically analyzed the trade-offs between accuracy, training time, and latency across different quantization strategies, providing essential guidance for deploying efficient deep learning systems in resource-constrained environments. This work, garnering 8 citations, has become a key reference for researchers and engineers optimizing AI workloads. Beyond this, Kong’s broader contributions span memory system design, security in embedded systems, and energy-efficient computing, making him a versatile figure in the field. His research not only advances academic understanding but also directly informs practical implementations in computer vision, natural language processing, and robotics, cementing his impact on both theory and application.
Research Focus
Key Achievements
Top Papers
- 1