STSIIML: Study on token shuffling under incomplete information based on machine learning

【Author】 Wang, Yilei; Li, Tao; Liu, Ming; Li, Chunmei; Wang, Hui

【Source】INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS

【影响因子】8.993

【Abstract】Transaction data on the public chain is open and transparent to all participants, which poses a potential threat to the privacy of participants. Some privacy-conscious token holders want to employ obfuscation methods to protect the origin and destination of their tokens, and the need for token shuffling services (TSS) arose at a historical moment. The prevailing token shuffling policies rely too heavily on blacklists in the token shuffling process, which is contrary to the idea of decentralization. Therefore, weakening or even eliminating the usage of blacklisting mechanisms in TSSs, which is an urgent issue to be addressed. In this paper, we adopt the idea of machine learning and propose a general framework for TSSs, which replaces the natural role under incomplete information with machine learning, so as to achieve the goal of eliminating the blacklist mechanism in TSSs. Then, the token shuffling process is constructed as an extended game under incomplete information on the basis of different token shuffling policies, and this incomplete information game is analyzed on the basis of poison policy, haircut policy, and suicide policy respectively. Finally, the sequential equilibrium under different games is investigated through simulations. The simulation results show that in the incomplete information game based on the poison policy, the participation of two players in the TSS is a sequential equilibrium, while in the incomplete information game based on the haircut policy, the players do not participate in the TSS is an sequential equilibrium.

【Keywords】blockchain; incomplete information game; machine learning; token shuffling service

【发表时间】

【收录时间】2022-09-06

【文献类型】实验仿真

【主题类别】

区块链治理-技术治理-异常/非法交易识别