Research on Reinforcement Learning Containerization Method Based on RISC-V Architecture
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
With the rapid advancement of artificial intelligence, reinforcement learning (RL) has shown great potential in intelligent decision-making, autonomous control, and adaptive systems. However, most existing RL frameworks rely on closed architectures such as x86 or ARM, which limit flexibility, scalability, and hardware customization. As an open-source instruction set architecture (ISA), RISC-V offers modularity, extensibility, and low power consumption, providing an ideal foundation for next-generation intelligent computing. This paper proposes a containerized RL framework based on RISC-V, aiming to build an efficient and portable environment for training and inference. The framework introduces a layered container architecture with instruction-level optimization, lightweight runtime, and fine-grained resource isolation. Experimental results on a RISC-V simulation platform demonstrate that the proposed method reduces startup latency and resource contention while improving computational efficiency and energy utilization, establishing a solid foundation for scalable AI deployment and RISC-V–based heterogeneous computing.
Fields of Study & Subjects
Parallel Computing and Optimization Techniques
Embedded Systems Design Techniques
Cloud Computing and Resource Management
Publication details
- Journal / source
- IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
- Pages
- 1532–1537
- Publication date
- 28 January 2026
- Language
- EN