Papers
1
Total Citations
59
H-Index
1
About
Youngwoo Kim is a hardware and computer architecture researcher whose work sits at the critical intersection of deep learning and energy-efficient computing. His most notable contribution centers on the design of specialized hardware accelerators for artificial intelligence workloads, particularly targeting the demanding requirements of deep reinforcement learning (RL) in mobile and autonomous systems. His 2019 paper, "A 2.1TFLOPS/W Mobile Deep RL Accelerator with Transposable PE Array and Experience Compression," which has garnered 59 citations, showcases his innovative approach to overcoming real-time processing constraints in action-control tasks — a challenge distinct from conventional recognition workloads. By introducing a transposable processing element array and experience compression techniques, Kim addressed the computational bottlenecks that prevent deploying deep RL on resource-constrained platforms like robots. His research is particularly significant for the robotics and autonomous systems communities, where bridging the gap between powerful AI algorithms and practical hardware limitations is paramount. Kim's work exemplifies the growing importance of co-designing algorithms and hardware to unlock the full potential of intelligent, real-time autonomous systems in energy-constrained environments.
Research Focus
Key Achievements
Top Papers
- 1