Sungpill Choi
Papers
1
Total Citations
59
H-Index
1
About
Sungpill Choi is a leading researcher in energy-efficient hardware accelerators for deep learning and reinforcement learning, with a focus on enabling real-time autonomous systems. His most cited work, "A 2.1TFLOPS/W Mobile Deep RL Accelerator with Transposable PE Array and Experience Compression" (2019, 59 citations), introduces a groundbreaking mobile accelerator that achieves exceptional energy efficiency—2.1 teraflops per watt—by combining a transposable processing element array with experience compression techniques. This innovation directly addresses the critical challenge of deploying deep reinforcement learning for action control in resource-constrained platforms like robots and drones, where real-time operation is paramount. Choi’s contributions bridge the gap between recognition and control tasks in neural networks, demonstrating how hardware design can optimize both computational throughput and power consumption. His work has been recognized for its practical impact on autonomous systems, offering a scalable solution for mobile and edge devices. With a growing citation footprint, Choi continues to influence the fields of deep learning hardware, reinforcement learning, and low-power computing, making him a notable figure in the advancement of intelligent, energy-efficient autonomous technologies.
Research Focus
Key Achievements
Top Papers
- 1