Stephen J. Rogers
Papers
1
Total Citations
3
H-Index
1
About
Stephen J. Rogers is a computer architect and systems researcher whose work focuses on optimizing memory performance in specialized computing platforms, particularly for computer vision and machine learning workloads. His most cited paper, "Exploiting architectural features of a computer vision platform towards reducing memory stalls" (2018), addresses a critical bottleneck in vision-based systems: memory access latency. By identifying and leveraging unique architectural features of vision platforms, Rogers demonstrates how to minimize memory stalls—a key factor limiting throughput in real-time applications like autonomous navigation and image recognition. Though early in his citation impact (3 citations for this work), his research contributes to the broader effort of bridging hardware-software co-design for efficient AI inference. Rogers’ approach is notable for its practical orientation, targeting tangible performance gains rather than theoretical abstraction. His work aligns with emerging trends in domain-specific architectures, where custom hardware accelerators are tailored to reduce data movement and energy consumption. For students and researchers exploring memory hierarchy optimization or vision system design, Rogers offers a grounded perspective on how architectural awareness can unlock performance in resource-constrained environments.
Research Focus
Key Achievements
Top Papers
- 1