George Kadianakis
@asn_d6
Followers
789
Following
16
Statuses
75
@soispoke @ralexstokes We could also do hybrid approaches where we use RANDAO to pick a big anonymity set, and then hold an election within it to pick the subset that will actually become includers. We might want to do this for performance reasons. This is closer to the model of
0
0
2
Similarly, if you are interested in implementing and benchmarking the SNARK-based LRS scheme from our post, drop us a message! We would love to get some prototypes out and start a formal specification of the scheme.
0
2
8
RT @benediktbuenz: Accumulation/Folding schemes are all the hype. However, all current schemes have proofs that are linear in the circuit s…
0
13
0
RT @RandomString00: Two interesting application of KZH-fold: A) Distributed PCD with low communication overhead, with concrete use cases…
0
3
0
You can find our research implementation in (major props to @RandomString00 for the heavy lifting!). We know there’s plenty of room for optimizations and code improvements, and we'd love for people to experiment with KZH-fold and push its limits!
0
1
6
@potuz_eth @ralexstokes IMO, as the number of devs/clients grows, taking major decisions in a zoom call with everyone inside just doesn't scale. The bad outcome here is that decisions will be taken in backrooms, while the inefficient ACD is used as a token of "decentralization".
1
0
4
@soubhik_deb WRT second question, Dave would run `GetSigners` and pick the heaviest merged signature (the one with the highest participation). Cheers!
0
0
9
@soubhik_deb Hey. The deferred expensive operation (the decider) in that scheme is essentially `Mergeverify`. As described in Appendix A, it's performed by anyone who wants to merge the sig forward, or by anyone who is just verifying the merged sig (potentially to include it in the block) 1/2
0
0
1
Our research highlights open research challenges related to the scalability of PCD constructions. Join us in pushing forward the field of distributed prover PCD by helping out with the open research questions!
0
0
1