Zero Knowledge
Latest recursive SNARKitecture with Izaak Meckler from O(1)Labs
Jan 22, 2020
Izaak Meckler from O(1) Labs discusses the launch of a recursive snark based incentivized test net, updates on the working SNARKitecture, and new use-cases for zkps in making computer programs more accountable to users.
51:49
Episode guests
AI Summary
AI Chapters
Episode notes
Podcast summary created with Snipd AI
Quick takeaways
- Snark technology has seen improvements in both implementation level and algorithmic level, resulting in faster computation and more efficient Snarks for applications beyond token transfers.
- Recursive snarks can be used to enforce specific algorithms in applications, ensuring transparency and accountability, such as in loan determinations, and have the potential to enable a more accountable and transparent digital society.
Deep dives
Improvements in Snark Technology
There have been significant improvements in both implementation level and algorithmic level in Snark technology. On the implementation side, a GPU prover has been developed, which is over four times faster than a 16-core machine. This has greatly improved the speed of Snark computation. Algorithmically, a new Snark construction is currently being implemented, which combines the AHP (Algebraic Holographic Proofs) from the Halo system with batching techniques. These advancements have made Snarks more efficient and have paved the way for applications beyond simple token transfers.
