The PhD project is quite flexible and should address one or more of the following aspects:
- Incorporating reliability and safety features into novel, low-cost, ultra-low-power, open-source RISC-V-based accelerators tailored for edge AI applications.
- Improving energy efficiency through architectural optimization, memory communication enhancements, and control strategies.
- Addressing lifetime and soft-error reliability through architectural and physical design methodologies, as well as system-level improvements.
- Developing an infrastructure for in-field fault management of RISC-V-based systems to prevent catastrophic system failures, including mechanisms for fault and aging detection, recovery, and using IJTAG for embedded instruments.
Given this context, which direction would you suggest I take when writing my PhD proposal? Additionally, what is the state-of-the-art in current research related to these topics?
Gaoqiang Qu
A PhD proposal on reliable RISC-V architecture for edge AI can focus on ensuring dependability, energy efficiency, and real-time performance in edge computing scenarios. Below is a structured outline of a possible proposal:
- To design and implement fault-tolerant mechanisms for RISC-V processors to mitigate errors in edge AI environments.
- To integrate hardware-software co-optimization strategies for efficient execution of AI workloads (e.g., neural network inference).
- To evaluate real-time performance enhancements in edge AI devices using customized RISC-V cores.
- To explore energy-efficient architectures for edge AI tasks without compromising reliability.
0 votes
0 thanks
Andi Ferataj
@Gaoqiang Qu
Those topinc looks basically same with my question, and so general topic, can make a examample to me, mybe also based or personal esperience?.
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training
More Andi Ferataj's questions See All
Similar questions and discussions