Saad Bin Nasir
Papers
2
Total Citations
116
H-Index
2
About
Saad Bin Nasir is a leading researcher in energy-efficient neuromorphic computing, with a focus on enabling autonomous intelligence in resource-constrained platforms. His work centers on mixed-signal time-domain architectures that bridge the gap between biological learning mechanisms and silicon hardware. Nasir’s most influential contribution is the development of a 55nm time-domain mixed-signal neuromorphic accelerator featuring stochastic synapses and embedded reinforcement learning, designed specifically for autonomous micro-robots. This work, cited 72 times, demonstrates how reinforcement learning—a bio-mimetic approach where agents learn from environmental rewards without human supervision—can be efficiently implemented in hardware. A related paper (44 citations) further explores the accelerator’s operation at sub-1V voltages, achieving an energy efficiency of 1.25 pJ/MAC. Together, these contributions represent a significant step toward true autonomy in intelligent agents, moving beyond inference-only systems to platforms capable of real-time learning. Nasir’s research is notable for its practical focus on low-power, compact designs that could enable next-generation autonomous systems, from microrobots to edge-AI devices.
Research Focus
Key Achievements
Top Papers
- 1
- 2