Lecture 10.3: What is a zk-SNARK?

[@quintinmakwe176]

Nothing short of GREAT. Succint explanations, yet to see any clearer introduction to zk-SNARKS, just great!!!!

[@smyasir0]

This is a BRILLIANT lecture ( especially for a Computer Engineer like me, whose head starts spinning when studying maths ) Very well explained !

[@itsshowtime6412]

Finally a real life use case of Zero Knowledge Proof. Good to see it being used within blockchains.

[@joshhoover1202]

Zero knowledge proof are already used all over the place, for example signature (such as ecdsa) is a zk proof. When you make the signature you are proving that you know the private key for an associated public key.

[@aminzakhirekar4168]

That was really great, simple yet comprehensive explanation of what is going on in the backstage of a zk-snark, thank you

[@GuyEshet]

Great and clear video, thanks a lot!

[@baldwincepeda943]

Thank you, great explanation!!!

[@yongkangchia1993]

what an aawesome simple explaination

[@kurtweinheimer5513]

Beautiful explanation thank you so much

[@mahdiech3284]

Amazing 😍👏

[@matinjiryaeisharahi4294]

that was amazing

[@tikaramsanyashi4927]

Nicely explained thanks

[@lherfel]

thanks for the zk explanation

[@scwyes]

Finally this makes sense to me thank you

[@gauravpoudel7288]

just got to 3:41 but I'm already liking this awesome explanation

[@BensDevJourney]

Finally, an actually good lecture on zk-Snarks... :)

[@michaelaziz5813]

great explanation of the argument system.....I really enjoyed the video although I expected more knowledge about the zk-snarks itself :(

[@lherfel]

thanks again

[@telotawa]

where do i find this 32:50 cs251

[@plusplus4591]

great

[@franklee8032]

But how does a proof look like? The Zokrates example at 24:37 leaves more questions than answers. We know that a proof has something to do with an arithmetic circuit, but not how it does so or why.

[@DavidWongTianyu]

Wondering why "soundness" is defined as "argument of knowledge". Also, why is bulletproof linear verification time? Also, I thought succinctness means polylog (not log)

[@taopaille-paille4992]

the statement to prove for Bulleproofs is of dimension N, where N is such that the message must be between 0 and 2^N -1. Bulletproofs just uses EC crypto,. No Polynomial commitments, no trusted setup. There are 2N such public generators (as opposed to 1 for Schnorr signatures, or 2 for Pedersen committments) and the proof verification in particular needs to multiply all these generators together. In total there are 2N + cste EC multiplications to do. The state of the art is Bulletproofs++ though, which uses a norm argument instead of a inner product argument like in Bulletproofs or Bulletproofs+. making it quite more efficient, although still with a linear verification time.

[@Enzoerb]

Have you seen the Midnight network?

[@lherfel]

zkp = a proof that shows one knows secret.... but anyone who does not know secret can produce the proof of knowing the secret?

[@2i2i39]

If Sim produces pi: is it then not possible for any 3rd party to run Sim without knowing w, thereby "proving" knowledge of w without actually knowing w.

[@jole0]

i think it would be a proof π and not the π, but im also interested to hear...

[@benjaminreynolds6064]

Same question. This seem inherently against security?

[@lherfel]

ditto! seems to contradict itself. zkp is giving proof that one knows secret.... but anyone who does not know secret can produce the proof???

[@vancao1965]

how did you do it can you share with me , thank you

[@zes3813]

no such thing as succinct x or remarx or taike or not etc, outx infix any nmw and any s perfect

[@chunghoang4605]

the video image is too poor, you need to fix it more

[@MRGCProductions20996]

23:50, can the witness be anything and it will just yield a proof? or does it need to be specifically chosen so that C(witness) = 0?

[@huytruongtien4847]

how did you do it can you share with me , thank you