Thangadurai Sivaram
Papers
1
Total Citations
72
H-Index
1
About
Thangadurai Sivaram is a pioneering researcher at the intersection of neuromorphic computing and mixed-signal integrated circuit design. His work focuses on creating energy-efficient hardware accelerators that mimic biological neural networks, enabling real-time learning in resource-constrained autonomous systems. His most cited paper, "A 55nm time-domain mixed-signal neuromorphic accelerator with stochastic synapses and embedded reinforcement learning for autonomous micro-robots" (2018, 72 citations), represents a landmark contribution to the field. This work introduced a novel time-domain processing approach that leverages stochastic synapses to implement reinforcement learning directly on-chip, eliminating the need for external training infrastructure. The accelerator's ability to support embedded learning while maintaining ultra-low power consumption makes it particularly suited for autonomous micro-robots operating in dynamic environments. Sivaram's innovations address a critical gap in neuromorphic hardware: enabling adaptive, real-time decision-making without cloud connectivity. His research has significant implications for edge AI, swarm robotics, and autonomous systems where power efficiency and on-device learning are paramount. By bridging the gap between neuroscience-inspired algorithms and practical silicon implementations, Sivaram is helping to shape the future of intelligent, self-learning machines.
Research Focus
Key Achievements
Top Papers
- 1