Yunxiang Zhang
Affiliation confirmed via AI analysis of OpenAlex, ORCID, and web sources.
Assistant Professor
Also affiliated: Binghamton University (2024–2025)
Research Areas
Links
Biography and Research Information
OverviewAI-generated summary
Yunxiang Zhang's research focuses on the design and optimization of hardware accelerators for neural network models, particularly those utilizing AdderNet architectures. His work investigates methods for improving efficiency and performance, such as activation-oriented quantization and memory optimization techniques like fused bias removal. Zhang has explored the application of these techniques to Field-Programmable Gate Arrays (FPGAs) for efficient acceleration. His recent publications also address operator-level neural architecture search for low-power and fault-tolerant Convolutional Neural Network (CNN) accelerators, considering factors like SRAM voltage scaling. Furthermore, he has investigated parameterized hardware generation for operations like summation-of-absolute-differences, relevant to AdderNet-based CNN models, and has a publication on dynamic partial reconfiguration based on FPGAs.
Metrics
- h-index: 2
- Publications: 6
- Citations: 11
Positions
-
Assistant Professor 2026–presentUniversity of Arkansas at Little Rock School of Engineering and Engineering Technology ORCID
Selected Publications
-
Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling (2026)
Collaboration Network
Top Collaborators
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
- Operator-Level Neural Architecture Search for Low-Power and Fault-Tolerant CNN Accelerators Under SRAM Voltage Scaling
Similar Researchers
Based on overlapping research topics