Anvesha Amaravati
Papers
2
Total Citations
55
H-Index
2
About
Anvesha Amaravati is a leading researcher in ultra-low-power, energy-efficient hardware for machine learning, with a primary focus on neuromorphic computing and embedded computer vision. Her work is pivotal in enabling intelligent, autonomous systems that operate under stringent power constraints, such as mobile robots and smart cameras. Dr. Amaravati’s most notable contribution is the development of a 55-nm, 1.0–0.4V, 1.25-pJ/MAC time-domain mixed-signal neuromorphic accelerator. This groundbreaking design, which has garnered 44 citations, introduces stochastic synapses to efficiently implement reinforcement learning, allowing mobile robots to learn from their environment without human supervision. This work directly addresses the challenge of bringing bio-mimetic, reward-based learning to resource-limited edge devices. Additionally, her research on a light-powered smart camera (11 citations) demonstrates a novel approach to gesture detection, achieving ultralow power consumption by extracting features directly from compressed sensor data. This innovation is critical for always-on, battery-less sensing applications. Through these achievements, Amaravati has established herself as a key innovator at the intersection of mixed-signal circuit design, neuromorphic engineering, and energy-autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2A Light-Powered Smart Camera With Compressed Domain Gesture Detection11 citations · 2017