Mathematician Jasper Zhang Introduces Proof of Sampling Protocol for Decentralized AI Networks
Mathematics prodigy Jasper Zhang, known for his gold medals at math olympiads in China and Russia, and his lightning-fast Ph.D. from UC Berkeley, is now turning his genius towards solving a critical problem at the crossroads of blockchain and AI.
Zhang, 28, is the leader of Hyperbolic, a cutting-edge startup focused on decentralized AI computing. The company announced on Thursday the launch of a groundbreaking protocol called “Proof of Sampling (PoSP),” designed to tackle trust issues in decentralized AI networks.
Co-founded in 2022 by Zhang and computer science Ph.D. Yuchen Jin from the University of Washington, Hyperbolic collaborated with researchers from Berkeley and Columbia University to develop PoSP. This innovative protocol combines math, computer science, and economics to incentivize integrity and reduce computational demands across decentralized networks.
In an interview with CoinDesk, Zhang described PoSP as the next evolution in decentralized network verification. The protocol can be applied not only to AI but also to layer-2 blockchains like rollups and actively validated services (AVSs) secured by restaking protocols.
A research paper detailing the Proof of Sampling Protocol, authored by Zhang and his team, was recently submitted to arXiv, a renowned open-access repository for scientific papers. The design of PoSP is based on a “pure strategy Nash Equilibrium,” a game theory concept popularized by mathematician John Nash, subject of the film “A Beautiful Mind.”
To complement the protocol, Hyperbolic is introducing “spML,” a specialized implementation of PoSP tailored for AI verification. According to Zhang, spML enhances verification speed, security, and economic feasibility.
As Hyperbolic prepares to put their innovative solutions to the test, the tech world eagerly awaits to see if Zhang’s mathematical prowess can revolutionize the intersection of blockchain and AI.
