PhD Defense: Practical Cryptography for Blockchains: Secure Protocols with Minimal Trust
Noemi Glaeser
Monday, October 28, 2024, 2:00-3:30 pm 
Abstract
In 2008, Satoshi Nakamoto introduced Bitcoin, the first digital currency without a trusted authority whose security is maintained by a decentralized blockchain. Since then, a plethora of decentralized applications have been proposed utilizing blockchains as a public bulletin board. This growing popularity has put pressure on the ecosystem to prioritize scalability at the expense of trustlessness and decentralization.
This work explores the role cryptography has to play in the blockchain ecosystem to improve performance while maintaining minimal trust and strong security guarantees. I discuss a new paradigm for scalability, called naysayer proofs, which sits between optimistic and zero-knowledge approaches. Next, I cover two on-chain applications: First, I show how to improve the security of a class of coin mixing protocols by giving a formal security treatment of the construction paradigm and patching the security of an existing, insecure protocol. Second, I show how to construct practical on-chain protocols for a large class of elections and auctions, which simultaneously offer fairness and non-interactivity without relying on a trusted third party. Finally, I look to the edges of the blockchain and formalize new design requirements for the problem of backing up high-value but rarely-used secret keys, such as those used to secure the reserves of a cryptocurrency exchange, and develop a protocol which efficiently meets these new challenges.
All of these works will be deployed in practice or have seen interest from practitioners. These examples show that advanced cryptography has the potential to meaningfully nudge the blockchain ecosystem towards increased security and reduced trust.
This work explores the role cryptography has to play in the blockchain ecosystem to improve performance while maintaining minimal trust and strong security guarantees. I discuss a new paradigm for scalability, called naysayer proofs, which sits between optimistic and zero-knowledge approaches. Next, I cover two on-chain applications: First, I show how to improve the security of a class of coin mixing protocols by giving a formal security treatment of the construction paradigm and patching the security of an existing, insecure protocol. Second, I show how to construct practical on-chain protocols for a large class of elections and auctions, which simultaneously offer fairness and non-interactivity without relying on a trusted third party. Finally, I look to the edges of the blockchain and formalize new design requirements for the problem of backing up high-value but rarely-used secret keys, such as those used to secure the reserves of a cryptocurrency exchange, and develop a protocol which efficiently meets these new challenges.
All of these works will be deployed in practice or have seen interest from practitioners. These examples show that advanced cryptography has the potential to meaningfully nudge the blockchain ecosystem towards increased security and reduced trust.
Bio
Noemi Glaeser is a Ph.D. Candidate in Computer Science at the University of Maryland and the Max Planck Institute for Security and Privacy advised by Jonathan Katz and Giulio Malavolta, respectively. Noemi is broadly interested in applied cryptography, especially in designing provably secure protocols to solve practical problems and to further user security and privacy. These interests have led her into blockchain research, where she has worked on the design and analysis of threshold wallets and on-chain protocols. Noemi was previously a research intern at NTT Research and a16z crypto and is a recipient of the NSF Graduate Research Fellowship.
This talk is organized by Migo Gui