Title:
Mechanism Design in Blockchain Protocols (link to video)
Speaker:
Matt Weinberg, Princeton
Abstract:
Blockchain protocols are often presented as specifying behavior that participants will take. Classical security analysis then partitions the players into 'honest' (who follow the specified behavior) and 'malicious' (who behave arbitrarily), and finds sufficient conditions on the fraction of honest players for a desirable outcome. However, blockchain protocols are really games: they provide a space of actions for players to take, and the protocol specifies a payoff for players as a function of the action profile.
In this talk, I'll give a survey of mechanism design challenges in blockchain protocol design. Some themes that will come up are:
- Incentive challenges in proof-of-work vs. proof-of-stake protocols
- Incentive challenges with block rewards vs. transaction fees
- Incentive challenges in longest-chain vs. byzantine consensus protocols
Time-permitting, I'll also try to highlight some themes that I find exciting about future challenges, such as: Maximal Extractable Value (MEV) and Trusted Randomness
Bio:
Matt is an assistant professor in Computer Science at Princeton. His main research area is Algorithmic Mechanism Design: the design of algorithms in settings with strategic players. A significant fraction of his research studies classical mechanism design settings (e.g. multi-item auctions) through an algorithmic lens. Since 2015, Matt has also been studying mechanism design in the blockchain space. Here, his research largely focuses on the incentives faced by participants in blockchain protocols, and understanding the extent to which they are incentivized to deviate from intended protocols.