- This event has passed.
Research Challenges in Computing
November 8 @ 11:00 am - 12:00 pm
Pratik Soni
Assistant Professor
November 8, 2024
11:00am
3147 MEB
https://utah.zoom.us/j/99963204158?pwd=tEpfDa4bQrzT7echP32lnt3LbY8fyP.1
Meeting ID: 999 6320 4158
Passcode: 595106
New Frontiers in Zero-Knowledge Proofs
Abstract: Zero-knowledge proofs allow a prover, holding a secret, to convince a verifier of some property of its secret, without revealing *any* additional information. Often described as merely theoretical constructs in cryptography, zero-knowledge proofs have recently received significant practical attention due to their prominence as a privacy-enhancing technology. These practical strides include novel proof systems with attractive (asymptotic) efficiency properties, emergence of software libraries that implement them in numerous languages, an industry-academic partnership for standardization, and novel hardware architectures that enable scalability. Despite these amazing advances, the focus has largely limited to scaling *general-purpose* proof systems, which oftentimes overfit applications or are inapplicable due to a mismatch in computational, data-access, or security threat models.
In this talk, I will summarize my recent works that take an *application-centric approach* to address these unmet needs. Firstly, I will introduce a novel application of zero-knowledge proofs for *provenance of digital media*, and describe opportunities for scaling such proofs by leveraging the structure of the provenance task. Secondly, I will introduce a new data-access model where the prover’s secret is *distributed* across a set of (possibly untrusting) low-memory and low-bandwidth devices. In such a distributed setting, I will highlight our recent work that makes progress on building truly distributed provers. Thirdly, I will extend the notion of *subversion security*, a critical post-Snowden realization, to zero-knowledge proof systems. Such subversion-secure proofs have applications to contingency payment systems – decentralized financial contracts that enable fair sale of digital assets. I will conclude with a few interdisciplinary questions for improving the security and scalability of zero-knowledge proof (implementations) and regulatory challenges they introduce in private decentralized finance.