Kimia Hassibi - Research
Enabling Computational Democratization: A Proof-of-Stake Bounty System for User-Proposed Problems and Solutions
In this work, we proposed a new cryptocurrency with a proof-of-stake consensus mechanism that allows miners to be rewarded for solving user-proposed problems. This cryptocurrency aims to incentivize miners of proof-of-work cryptocurrencies to use their compute power to do useful work. This research has been awarded Best Paper (1st Place) at the 2023 IEEE MIT Undergraduate Research Technology Conference, and it is set to be published in the IEEE Xplore database.
Markov Chain Quantal-Response
Real-world players in games are susceptible to making errors. In the quantal-response model, the players perceive utilities that include random errors. Here, the players' responses are probabilistic, with players having a higher likelihood of choosing better responses than worse responses. We study the dynamics of the quantal-response model by analyzing the stationary distribution of a Markov chain with a transition matrix defined by quantal-responses. We consider a game in which the players are divided into two groups, and the utility of a player in one group depends only on the actions of the players in the other group. We find that if we allow all the players in such a game to simultaneously revise their responses, the stationary distribution is the product of the marginal distributions of each group.