Private Access Control for Function Secret Sharing
Sacha Servan-Schreiber, Simon Beyzerov, Eli Yablon, Hyojae Park
In function secret sharing (FSS) a dealer can privately outsource some function to a set of evaluators. These evaluators can evaluate the function on some input without learning the result of the computation . We introduce the notion of private access control for function secret sharing, whereby we restrict the dealer to only be able to secret-share a function that they have been granted access to, all while maintaining privacy. Written in conjunction with MIT PRIMES. Cryptography paper developing distributed zero knowledge proof protocols towards access control for secret-shared functions.
Cloak: A Versatile Framework for Anonymous Authentication
AuthorsSacha Servan-Schreiber, Simon Beyzerov, Eli Yablon
Many online resources are pay-walled or otherwise restricted to the public. To access these resources, users need to prove access privileges through an authentication mechanism. For example, news websites such as the New York Times (NYT) offer news articles exclusively to paid subscribers. To access NYT articles, users provide the NYT website with an email and password associated with an active subscription (users to the NYT). What if users want to preserve their privacy and not reveal which articles they consume? An anonymous authentication mechanism can solve this problem. We introduce a novel zero-knowledge proof protocol towards the development of a framework for anonymous authentication. Written in conjunction with MIT PRIMES. Cryptography paper developing privacy preserving authentication through secret-shared zero-knowledge proof protocols.
Authenticated Metadata-hiding Anonymous Communication
Paired with the rise of consumer data analytics, the continued transition of interpersonal communication onto the Internet has made privacy an increasingly important requirement in many network systems. Although attempts by services such as WhatsApp, iMessage, and Facebook Messenger to employ methods of end-to-end encryption have succeeded in securing message content, leaked metadata has exposed the power of underlying communication data. In this work, we present a set of lightweight constructions for anonymous communication with cryptographic privacy guarantees for users. Independent research paper exploring how anonymous authentication techniques can be used to improve the current state of the art in anonymous communication systems with cryptographic privacy guarantees.
Seed-Homomorphic Pseudorandom Generators from Learning with Errors / Multi-Server Distributed Point Functions
(Still in the works!)
Moderately Hard Functions
(Still in the works!)
Mathematical modeling paper written during a the 2022 International Mathematical Modeling Competition (IMMC). Modeled planar loading and disembarking while optimizing both processes in a diverse set of aircrafts.
Mathematical modeling paper written during the 2021-22 High School Mathematical Contest in Modeling (HiMCM). Modeled the causes, future outlook, and potential consequences of rapidly depleting water levels at Lake Mead, Nevada, the nations largest freshwater reservoir.
Mathematical modeling paper written during the 2022 MathWorks Mathematical Modeling Competition. Modeled changes in employment caused by a pandemic-induced virtual workplace. Considered future outlook, causes, and conducted a broader mathematical and social analysis.
Coding and more
Simon Beyzerov, Aaron Tian
Mealgo: Recipe Filtering
Android application designed to provide intelligent meal and recipe recommendation that considers both pricing and individual user preferences. Developed with a tailored dataset of over 500,000 recipes and thousands of unique ingredients, all with their respective nutritional information. Developed in Kotlin.
NaVi: Aid for Independent Navigation for the Visually Impaired
Designed a physical aid for the independent navigation of public spaces for those with visual impairments. Designed software for automated guidance and designed a haptic device to aid those attempting to navigate through unadapted environments. Allows for independent parties to deploy schematics that aid individuals in the respective space.
Mass Academy Coursework
For prospective Mass Academy students, alumni, or for anyone interested: past assignments and projects completed during the 2021-22 school year are available upon request. Certain assignments will be available in the future.