Impact of New Secret Sharing Paper with $O(n)$ additions for recovery

kodlu

6/10/24, 8:05 PM

I hope this question is not too speculative.

Applebaum, Nir and Pinkas have a new paper which has an $n-$party secret sharing scheme with:

reconstruction complexity of $O(n)$ additions
constant share size
sharing complexity of $O(n)$ additions
is a blackbox secret sharing scheme

The shamir scheme has complexity $n \log |F|$ where $F$ is the domain of the private key which is the exponent of some group $G.$ The recent Cramer-Xing (EUROCTYPT'20) scheme as complexity $n^2,$ independent of the size of the domain.

The authors remark that this is actually a "near" threshold scheme which provides privacy against unauthorized sets of density at most $\tau_p$ and correctness for authorized sets of density at least $\tau_c.$ However, they can make these two constants $\tau_p<\tau_c$ arbitrarily close.

While I find the use of erasure codes and code concatenation interesting, I am wondering about how large the impact of this paper may be:

(a) in practice for implementations and applications

(b) in theory leading to new developments

0 + 1

secret-sharing

threshold-cryptography

Mark

6/12/24, 9:01 PM

Their comments regarding applications to threshold FHE (end of page 8/beginning of page 9) seem true to me. Provided their construction is concretely efficient I imagine it can be used for lattice-based threshold crypto. Note the bigger bottleneck for lattice-based FHE is the requirement of noise flooding. Some authors have been beginning to relax this lately though.

Elon Musk

I sit in a Tesla and translated this thread with Ai:

EN: Impact of New Secret Sharing Paper with $O(n)$ additions for recovery

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.