This week, Arantxa Zapico of the Ethereum Foundation presents new work (co-authored with Vitalik Buterin, Dmitry Khovratovich, Mary Maller, Anca Nitulescu, and Mark Simkin) called Caulk, which examines position-hiding linkability for vector commitment schemes. One can prove in zero knowledge that one or more values that comprise commitment cm all belong to the vector of size committed to in C. Caulk can be used for membership proofs and lookup arguments and outperforms all existing alternatives in prover time by orders of magnitude.
https://eprint.iacr.org/2022/621
[2024]Digital Global Overview Report 2024 Meltwater.pdf
Caulk: zkStudyClub: Caulk - Lookup Arguments in Sublinear Time (A. Zapico)
1. Caulk: Lookup
Arguments
in Sublinear Time
Vitalik Buterin
Dmitry Khovratovich
Arantxa Zapico Mary Maller Anca Nitulescu
Mark Simkin
Universitat Pompeu Fabra Ethereum Foundation Protocol Labs
14. STATE OF THE ART
Transparent setup
Linear prover and
verifier
Discrete-log
15. STATE OF THE ART
Transparent setup
Linear prover and
verifier
RSA Accumulators
Discrete-log
16. STATE OF THE ART
Transparent setup
Linear prover and
verifier
RSA Accumulators
Discrete-log
Constant prover
Trusted parameters
17. STATE OF THE ART
Transparent setup
Linear prover and
verifier
Merkle Trees
RSA Accumulators
Discrete-log
Constant prover
Trusted parameters
18. STATE OF THE ART
Transparent setup
Linear prover and
verifier
Merkle Trees
RSA Accumulators
Discrete-log
Transparent setup
Need a zkSNARK on top
Constant prover
Trusted parameters
19. STATE OF THE ART
Pairing-based
Transparent setup
Linear prover and
verifier
Merkle Trees
RSA Accumulators
Discrete-log
Transparent setup
Need a zkSNARK on top
Constant prover
Trusted parameters
20. STATE OF THE ART
Pairing-based
Transparent setup
Linear prover and
verifier
Merkle Trees
RSA Accumulators
Discrete-log
Constant proof +
verifier
Linear prover
Transparent setup
Need a zkSNARK on top
Constant prover
Trusted parameters
36. ROOTS OF UNITY
1.
2.
Sparse Lagrange and vanishing polynomials
Any u such that uN
=1 is an Nth root of unity
zH
(X)=XN
-1 λi
(X)= (⍵i-1
(XN
-1)) ((X-⍵i-1
)N)-1
H={1,⍵,⍵2
,…,⍵N-1
}, ⍵N
=1
37. ROOTS OF UNITY
1.
2.
Sparse Lagrange and vanishing polynomials
Any u such that uN
=1 is an Nth root of unity
zH
(X)=XN
-1 λi
(X)= (⍵i-1
(XN
-1)) ((X-⍵i-1
)N)-1
H={1,⍵,⍵2
,…,⍵N-1
}, ⍵N
=1
If uN
=1 u=⍵something
74. [z]=[a(x-⍵i-1
)]
1
2
3
4
Construct f(X) of degree log(N)+6,
f(X)=Σj
fj
µj
(X)
Well formation
New set of roots of unity! V
Lagrange polynomials {µj
(X)}
m=1
Prove [z]=[ax+b] (b=a⍵i-1
)
75. [z]=[a(x-⍵i-1
)]
1
2
3
4
Construct f(X) of degree log(N)+6,
f(X)=Σj
fj
µj
(X)
Well formation
Prove f5
=b/a, and for j = 6,...,log(N)+5 fj
=fj-1
fj-1
m=1
Prove [z]=[ax+b] (b=a⍵i-1
)
76. [z]=[a(x-⍵i-1
)]
1
2
3
4
Construct f(X) of degree log(N)+6,
f(X)=Σj
fj
µj
(X)
Well formation
Prove f5
=b/a, and for j = 6,...,log(N)+5 fj
=fj-1
fj-1
Should be ⍵i-1
m=1
Prove [z]=[ax+b] (b=a⍵i-1
)
77. [z]=[a(x-⍵i-1
)]
1
2
3
4
Construct f(X) of degree log(N)+6,
f(X)=Σj
fj
µj
(X)
Well formation
Should be ⍵i-1
f5+j
is the 2j
th power
of ⍵i-1
m=1
Prove [z]=[ax+b] (b=a⍵i-1
)
Prove f5
=b/a, and for j = 6,...,log(N)+5 fj
=fj-1
fj-1
78. [z]=[a(x-⍵i-1
)]
1
2
3
4
Construct f(X) of degree log(N)+6,
f(X)=Σj
fj
µj
(X)
Well formation
Prove flog(N)+5
=1
m=1
Prove [z]=[ax+b] (b=a⍵i-1
)
Prove f5
=b/a, and for j = 6,...,log(N)+5 fj
=fj-1
fj-1
79. [z]=[a(x-⍵i-1
)]
1
2
3
4
Construct f(X) of degree log(N)+6,
f(X)=Σj
fj
µj
(X)
Well formation
Prove flog(N)+5
=1
(b/a)N
=1 !!!
Prove [z]=[ax+b] (b=a⍵i-1
)
m=1
Prove f5
=b/a, and for j = 6,...,log(N)+5 fj
=fj-1
fj-1
114. CREDITS: This presentation template was
created by Slidesgo,including icons by
Flaticon,infographics & images by Freepik
THANKS!
https://eprint.iacr.org/2022/621