Andrew DeCandia
Papers
1
Total Citations
2
H-Index
1
About
Andrew DeCandia is a rising researcher at the frontier of neuromorphic and in-memory computing, with a focus on hardware-software co-design for energy-efficient artificial intelligence. His most cited work, "In-Memory Computation Using CMOS-Integrated Resistive RAM for Robotic Navigation" (2024), demonstrates a pivotal contribution: the integration of Resistive Random Access Memory (ReRAM) into a CMOS platform to perform vector-matrix multiplication (VMM) directly in memory. This approach dramatically reduces the energy overhead of neural network inference, enabling real-time, low-power robotic navigation. By tackling the von Neumann bottleneck, DeCandia’s research shows how ReRAM-based in-memory computing can make autonomous systems—from drones to mobile robots—both faster and more efficient. Though early in his career, his work has already garnered attention for its practical hardware implementation, bridging the gap between emerging memory technologies and real-world AI applications. His achievements point toward a future where energy-constrained edge devices can run complex neural networks without sacrificing performance.
Research Focus
Key Achievements
Top Papers
- 1