Lionel Jure
Papers
1
Total Citations
34
H-Index
1
About
Lionel Jure is a leading researcher in energy-efficient artificial intelligence hardware, specializing in the design of ultra-low-power AI-enabled Internet-of-Things (AI-IoT) systems-on-chip (SoCs). His work focuses on bridging the gap between compute-intensive deep neural network (DNN) inference and the stringent power constraints of emerging applications like augmented reality, personalized healthcare, and nano-robotics. Jure’s most cited paper, a 2023 ISSCC publication with 34 citations, introduces a groundbreaking AI-IoT SoC that achieves 12.4 TOPS/W at 136 GOPS—a record efficiency for its class. This chip integrates 16 RISC-V cores with precision-scalable DNN acceleration supporting 2-to-8-bit operations, alongside adaptive body biasing that boosts performance by 30%. His contributions demonstrate how heterogeneous architectures and voltage scaling can enable complex AI tasks within a few tens of milliwatts, directly impacting the viability of edge intelligence. Jure’s work is widely recognized for advancing the frontier of near-threshold computing and mixed-precision neural networks, making him a key figure in the push toward autonomous, battery-powered AI devices.
Research Focus
Key Achievements
Top Papers
- 1