Non power of 2 Inner-product argument

Question

Non power of 2 Inner-product argument

suyash67 opened this issue 5 years ago · comments

Hi @lovesh. Can you help me out in figuring out how to go about using inner product argument for input vectors of sizes other than powers of 2?

Lovesh Harchandani · Answer 1 · Tue Feb 04 2020 22:17:26 GMT+0800 (China Standard Time)

Hi @suyash67, sorry for the delay. You can safely pad with 0s to make the vectors of size power of 2. That will increase the size of L and R by 1. The commitment P would not change since the elements of the vectors are either as exponents or part of inner product. Added a test showing this.

Suyash Bagad · Answer 2 · Wed Feb 05 2020 23:24:10 GMT+0800 (China Standard Time)

Hey! Thanks a lot for your response and help. I am working with a different elliptic curve group. I will try the approach you suggested. Hope it would work. Will let you know if I need any further assistance! Thanks again, *Regards,* *Suyash Bagad,*

…

On Tue, 4 Feb 2020 at 19:47, Lovesh Harchandani ***@***.***> wrote: Hi @suyash67 <https://github.com/suyash67>, sorry for the delay. You can safely pad with 0s to make the vectors of size power of 2. That will increase the size of L and R by 1. The commitment P would not change since the elements of the vectors are either as exponents or part of inner product. Added a test <https://github.com/lovesh/bulletproofs-amcl/blob/master/src/ipp.rs#L394> showing this. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1?email_source=notifications&email_token=AEVWN5NSB3FZSLWOFSJYF3DRBF2HNA5CNFSM4JZ3FT4KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKXYXKI#issuecomment-581929897>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEVWN5OFMVIQIF3QDMWGFOLRBF2HNANCNFSM4JZ3FT4A> .

Lovesh Harchandani · Answer 3 · Wed Feb 05 2020 23:57:41 GMT+0800 (China Standard Time)

You are welcome. May I ask which curve you are using?

Suyash Bagad · Answer 4 · Thu Feb 06 2020 00:13:06 GMT+0800 (China Standard Time)

I am working with secp256k1 curve using KZen Network's elliptic curve cryptography package. Regards, Suyash Bagad Final year Dual Degree Student, EE, IIT Bombay

…

On Wed, 5 Feb 2020, 9:27 pm Lovesh Harchandani, ***@***.***> wrote: You are welcome. May I ask which curve you are using? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1?email_source=notifications&email_token=AEVWN5J4PLGCAZFHEIJ3M53RBLOXNA5CNFSM4JZ3FT4KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEK36GYQ#issuecomment-582476642>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEVWN5L2ZODBS7CZZ6SAAI3RBLOXNANCNFSM4JZ3FT4A> .

Lovesh Harchandani · Answer 5 · Thu Feb 06 2020 00:24:58 GMT+0800 (China Standard Time)

That seems nice. Btw, just wanted to make you aware that this wrapper can be used with secp curve by compiling with the secp256k1 feature. Check here for more curves https://github.com/lovesh/bulletproofs-amcl/blob/master/Cargo.toml#L22

Suyash Bagad · Answer 6 · Thu Feb 06 2020 02:45:27 GMT+0800 (China Standard Time)

That's great! Will try out. Regards, Suyash Bagad Final year Dual Degree Student, EE, IIT Bombay

…

On Wed, 5 Feb 2020, 9:54 pm Lovesh Harchandani, ***@***.***> wrote: That seems nice. Btw, just wanted to make you aware that this wrapper can be used with secp curve by compiling with the secp256k1 feature. Check here for more curves https://github.com/lovesh/bulletproofs-amcl/blob/master/Cargo.toml#L22 — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1?email_source=notifications&email_token=AEVWN5LGZF5U2M4YXYXDMCDRBLR5XA5CNFSM4JZ3FT4KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEK4BOQY#issuecomment-582489923>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEVWN5M3AD3DH5HFP7HR4ILRBLR5XANCNFSM4JZ3FT4A> .