Distributed caching method and system integrating reinforcement learning and collaborative learning to prove consensus

By integrating reinforcement learning and collaborative learning to prove consensus in the Web3.0 cache system, the shortcomings in performance, security and decentralization in the existing technology are solved, and an efficient, secure and decentralized cache system is realized.

CN119201996BActive Publication Date: 2025-03-25NANJING UNIV OF SCI & TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202411228645.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2025-03-25
Estimated Expiration
2044-09-03

AI Technical Summary

Technical Problem

The existing Web3.0 caching system has shortcomings in performance, security and decentralization. Traditional caching strategies cannot meet their needs and pose a risk of centralization.

Method used

A distributed cache method that integrates reinforcement learning and collaborative learning to prove consensus is adopted, and a distributed cache framework is established through multi-agent reinforcement learning to realize an indirect cache hit model based on reinforcement learning network, and a single point of failure risk of cache nodes is reduced through collaborative learning to prove consensus.

Benefits of technology

It improves the performance and security of the cache system, meets the decentralized needs of Web3.0, reduces the risk of single point failure of cache nodes, and improves the cache hit rate.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention belongs to the field of distributed caching, and discloses a distributed caching method and system that integrates reinforcement learning and collaborative learning to prove consensus, the method comprising: a user node initiates a request to an oracle node through an application program interface; the oracle node broadcasts a preparation message; the oracle node trains an indirect cache hit model based on a reinforcement learning network; the oracle node broadcasts the model after training; the oracle node verifies the model after training; the oracle node caches the required content and responds to the user request, the cached required content includes a direct cache hit based on a local cache pool and an indirect cache hit based on a reinforcement learning network; the present invention improves the centralized training distributed execution paradigm, ensures the effectiveness of training, and thus achieves good multi-agent cooperation performance. The distributed caching framework proposed by the present invention is suitable for a variety of Web3.0 scenarios, and improves the security and reliability of the distributed caching framework.
Need to check novelty before this filing date? Find Prior Art