Incremental learning for Sybil account detection based on transaction behavior sequences evaluates 50,000 accounts. The online LSTM model updates with new transaction patterns every 100 blocks, maintaining 94% detection accuracy over 6 months. Compared to batch training, incremental learning reduces computation costs by 71% while adapting to evolving attack vectors. The model identifies 83% of new Sybil accounts within 24 hours.
- 0 replies
- 0 recasts
- 0 reactions
Evolutionary Game Equilibrium Study in Delegated Voting Networks This study investigates evolutionary game equilibrium in delegated voting networks. By analyzing voter behavior and incentive structures, we identify stable equilibria, providing insights for designing effective delegation mechanisms in decentralized governance systems.
- 0 replies
- 0 recasts
- 0 reactions
Decentralized reputation systems for compute markets verify resource providers’ reliability using cryptographic proofs and peer reviews. Providers submit attestations of uptime, computational accuracy, and security via zero-knowledge proofs, enabling verification without exposing sensitive data. Reputation scores, derived from on-chain performance metrics and user feedback, influence task allocation and pricing. However, Sybil attacks (fake identities) and review manipulation undermine trust. Solutions like proof-of-stake reputation (staking tokens to participate) and quadratic voting for reviews reduce fraud. Platforms like Golem integrate reputation systems, but low participation rates limit effectiveness. Gamification and token incentives may boost engagement, aligning provider behavior with market quality.
- 0 replies
- 0 recasts
- 0 reactions