Meta Reinforcement Learning Based Dynamic Tuning for Blockchain Systems in Diverse Network Environments
This program is tentative and subject to change.
The evolution of blockchain technology across various areas has highlighted the importance of optimizing blockchain systems’ performance, especially in fluctuating network bandwidth conditions. We observed that the performance of blockchain systems exhibits variations, and the optimal parameter configuration shifts accordingly when changes in network bandwidth occur. Current methods in blockchain optimization require establishing fixed mappings between various environments and their optimal parameters. However, this process exhibits poor sample efficiency and lacks the ability for fast adaptation to novel bandwidth environments. In this paper, we propose MetaTune, a meta-reinforcement-learning based dynamic adaptation method for blockchain systems. MetaTune can quickly adapt to unknown bandwidth changes and automatically configure optimized parameters. Through empirical evaluations of a real-world blockchain system, ChainMaker, we demonstrate that MetaTune significantly reduces the training samples needed for generalization across different bandwidth environments compared to non-adaptive methods. Our findings suggest that MetaTune offers a promising approach for efficiently optimizing blockchain systems in dynamic network environments.
This program is tentative and subject to change.
Sat 27 JulDisplayed time zone: Beijing, Chongqing, Hong Kong, Urumqi change
11:00 - 12:15 | Session 4 - Network & ConsensusResearch Track at Main Room Chair(s): Yury Yanovich Skolkovo Institute of Science and Technology; Faculty of Computer Science, HSE University | ||
11:00 15mPaper | IBFT: An Impartial Byzantine Fault Tolerance Consensus Protocol for Blockchain Research Track Yangpu Zeng Zhejiang Normal University, Feilong Lin Zhejiang Normal University, Lei Tian Zhejiang Normal University, Jiahao Gan Zhejiang Normal University, Zhongyu Chen Zhejiang Normal University | ||
11:15 15mPaper | Fault Tolerance Testing and Tuning for Consortium Blockchain Research Track Taiwu Pang East China Normal University, Zheming Ye East China Normal University, Zhao Zhang East China Normal University, Cheqing Jin East China Normal University | ||
11:30 15mPaper | ATBFT-Automatically switch consensus protocol Research Track Yuxuan Lu School of Software, Shandong University, Jinan 250101, PR China, Chang Liu School of Computing Science, Newcastle University, Newcastle NE1 7RU, PR United Kingdom, Lanju Kong Shangdong University, Xiangyu Niu School of Software, Shandong University, Jinan 250101, PR China | ||
11:45 15mPaper | An Efficient Bitcoin Network Topology Discovery Algorithm for Dynamic Display Research Track Zening Zhao Tianjin University of Technology, Jinsong Wang Tianjin University of Technology, Miao Yang Tianjin University of Technology, Haitao Wang Tianjin University of Technology | ||
12:00 15mPaper | Meta Reinforcement Learning Based Dynamic Tuning for Blockchain Systems in Diverse Network Environments Research Track Yue Pei Beihang University, Mengxiao Zhu North China University of Technology, Chen Zhu Beihang University, weihusong Beihang University, Yi Sun Chinese Academy of Sciences, Lei Li Zhongguancun Laboratory, Haogang Zhu Beihang University | ||