TSHML: Token Shuffling under Haircut Policy Based on Machine Learning

Abstract

The token shuffling policy determines whether the shuffling parties have shuffling motives. Existing token shuffling policies, such as poisoning policy, haircut policy, and suicide policy, all need to rely on blacklists in the implementation process. Blacklisting itself runs counter to the idea of decentralization. Therefore, how to weaken the blacklist in the token shuffling policy is an urgent problem to be solved. Utilizing the concept of machine learning, this paper proposes a general framework for token shuffle service under incomplete information, mainly Leveraging machine learning to replace the role of nature under incomplete information. Then, according to the haircut policy, the token shuffling service is reduced to an incomplete information game model. Finally, the sequential equilibrium under the incomplete information game based on haircut policy is analyzed by simulation. The simulation results show that in the incomplete information game based on haircut policy, it is an sequential equilibrium that players do not shuffle their tokens.