Designing Game theoretically Sound Fair and Private Multi agent Systems

Abstract

Multi-agent systems (MAS) are distributed systems composed of multiple autonomous newlineagents interacting to achieve a common or conflicting goal. MAS tackles complex and newlinedynamic problems that a single agent cannot solve, resulting in better problem-solving newlineskills, enhanced reliability, and improved scalability. This thesis explores the challenges newlinefacing MAS, particularly related to their game-theoretic, fairness, security, and privacy newlineguarantees. newlineA game-theoretically sound MAS is one where the agent interaction can be modeled newlineas a game and analyzed using game-theoretic concepts. This leads to a more stable and newlineefficient system, as agents are incentivized to make decisions that align with the system newlinegoals. This thesis focuses on civic crowdfunding, a method for raising funds through newlinevoluntary contributions for public projects (e.g., public parks). Our work enriches the newlineexisting literature by designing more inclusive mechanisms and providing fairer rewards newlineand efficiency over the blockchain. newlineFairness is also an essential aspect of MAS as it ensures that the actions and outcomes newlineof agents are equitable and just, resulting in MAS s long-term stability and sustainability. newlineThis thesis looks at fair incentives in Transaction Fee Mechanisms (TFM). Blockchains newlineemploy TFMs to include transactions from the set of outstanding transactions in a block. newlineWe argue that existing TFMs incentives are misaligned for a cryptocurrency s greater newlinemarket adoption. We propose TFMs that provide fairer rewards to the transaction creators newlineand minimize the surplus collected to the creators. newlinevii newlineviii newlineLast, security and privacy are crucial aspects of MAS, as the autonomy and decentralization newlineof agents in MAS can lead to the exposure of sensitive information. In this thesis, newlinewe specifically focus on privacy guarantees for MAS like (i) auctions, (ii) voting, and (iii) newlinedistributed constraint optimization (DCOPs). We propose privacy-preserving applications newlinethat preserve agents sensitive information while proving the computation s verifiability