Mechanism Design for Decentralized Systems
Blockchains offer desirable properties such as transparency, immutability, verifiable randomness, and resistance to monopolization—features that are difficult to realize in centralized systems. However, their decentralized nature poses a unique challenge for mechanism designers: the miners who maintain these systems are not inherently trustworthy. Unlike traditional mechanisms, which often rely on a trusted party such as an auctioneer or a government agency, blockchain miners are self-interested and may deviate from the prescribed mechanism when doing so is profitable.
In this work, we investigate both the challenges and opportunities of decentralized mechanism design. First, we examine how minor deviation introduces new obstacles. In blockchain systems, users participate in transaction fee mechanisms (TFMs), which determine transaction inclusion and payments. The blockchain community has long sought a mechanism that is incentive-compatible not only for users but also in the presence of collusion between miners and users. We prove that such a mechanism is fun damentally impossible. However, this impossibility result can be circumvented through cryptographic techniques or by rethinking the notion of incentive compatibility.
Second, we explore how blockchains can improve traditional mechanism design. Modern digital mar ketplaces typically rely on centralized platforms, such as eBay or Google, to mediate transactions. These platforms control all communication between buyers and sellers and can manipulate auctions for their own benefit. We provide a complete characterization of platform-assisted auctions when the platform is strategic and may collude with buyers and sellers. Further, we show how blockchains, functioning as a public bulletin board, can facilitate the design of incentive-compatible platform-assisted auctions.
Funding
Collaborative Research: SaTC: CORE: Medium: Game Theory, Economics, and Mechanism Design for Blockchains
Directorate for Computer & Information Science & Engineering
Find out more...NSF-BSF: SaTC: CORE: Small: Secure Massively Parallel Computations: Foundations and Constructions
Directorate for Computer & Information Science & Engineering
Find out more...SaTC: CORE: Large: Viaduct: A Framework for Automatically Synthesizing Cryptographic Protocols
Directorate for Computer & Information Science & Engineering
Find out more...History
Date
2025-04-28Degree Type
- Dissertation
Department
- Electrical and Computer Engineering
Degree Name
- Doctor of Philosophy (PhD)