Aayush Ankit
Papers
1
Total Citations
41
H-Index
1
About
Aayush Ankit is a leading researcher at the forefront of hardware-software co-design for energy-efficient machine learning, with a primary focus on in-memory computing (IMC) architectures. His most-cited work, "Circuits and Architectures for In-Memory Computing-Based Machine Learning Accelerators" (2020, 41 citations), addresses the critical bottleneck of data movement in deep neural network (DNN) accelerators. Ankit’s key contribution lies in pioneering circuit and architectural techniques that enable matrix-vector multiplications directly within memory arrays, dramatically reducing energy consumption and latency compared to traditional digital accelerators. By bridging the gap between emerging memory technologies and practical DNN workloads, his research has helped lay the groundwork for next-generation edge AI hardware. Ankit’s work is particularly notable for its holistic approach—spanning from device-level circuit design to system-level performance analysis—making it highly influential among both academic researchers and industry practitioners seeking to deploy efficient AI in resource-constrained environments. His contributions continue to shape the trajectory of non-von Neumann computing for machine learning.
Research Focus
Key Achievements
Top Papers
- 1