6 ms·
Lurk – Language for Recursive ZK-SNARKs Inspired by Common Lisp and Scheme
- diggan 4y agoMultiple interesting features: - Lurk program execution can be proved in zero knowledge. - Lurk proofs support multiple backend SNARK proving systems. - Lurk enables incremental computation and proofs in unbounded loops. - Lurk provides conditional control flow. - Lurk programs are data and vice versa. - Lurk data is content-addressable for compatibility with IPLD/IPFS. That it's a lisp just makes it all better as well. Here's a Fibonacci example: ;; (FIB TARGET) computes the element of the Fibonacci sequence at TARGET (zero-indexed). (letrec ((next (lambda (a b n target) (if (eq n target) a (next b (+ a b) (+ 1 n) target)))) (fib (next 0 1 0))) (fib 1))
- tylersmith 4y agoGenerally recursive here means that the program is verifying previous proofs from the program. I'm trying to understand how this fits in here wrt to the recursive function.
- porcuquine 4y agoIt's confusing because there are two distinct 'recursive things' involved. You may be noting that distinction. On the one hand, Lurk allows expression of recursive computations (like in the cited example). This doesn't necessarily demand 'verifying previous proofs' (in proofs). However, recursive SNARKs (where proofs of proofs are verified ad infinitum) are useful for actually producing a single succinct proof of a potentially unbounded recursive computation. So there is a practical relationship between the two recursive aspects, but they're not strictly the same thing. This is indeed a bit of a confusing subtlety.
- noman-land 4y agoSo you could do a collaborative computation of a particular index in fibonacci by taking turns computing up to a certain index, getting the hash and handing it off like a baton to someone else who could prove the starting value they're given is correct for the starting index and then keep going?
- porcuquine 4y agoYep, that's a perfect minimal example of the kind of thing we could do. Exactly how to 'extend' the computation depends on how it's set up. What you describe is possible, as are more elaborate applications in which the 'result so far' includes a log of all previous results (basically a memoized table of the recursive computation so far). If you have a proof that the merkle root of this table is correct, then you can just look up any so-far-known value. Taken to its extreme, this could be used to produce a collective database reducing arbitrary deterministic functions to lookup — what I think of as 'compute-addressable' (riffing on content-addressable) data. Since Lurk can express arbitrary computation in its own data language, such a universal database wouldn't strictly require per-function circuits (though for performance, some customization will help).
- noman-land 4y agoSo incredibly cool. I've been thinking about a use case for the 'result so far' mechanic and I've been trying to figure out how to build it. Perhaps I'll give lurk a try :). Are you on any zk chat channels on matrix/telegram/discord/etc? Would love to chat more about this.
- tylersmith 4y agoThe term you want to look into is Incrementally Verifiable Computation.
- orthecreedence 4y agoI wonder how this compares to something like zexe (https://github.com/brucechin/zexe https://github.com/brucechin/zexe). This domain interests me although I have little practical knowledge/experience in it.
- porcuquine 4y agoZEXE has one layer of recursive composition, so you can effectively verify proofs in a proof – but not proofs of those proofs. Lurk initially targets Nova for unbounded recursion.
- porcuquine 4y agoHi, this is my project, though I wasn't expecting to see it here just now. I gave some quick (probably too quick) answers below, and I'm happy to try to answer further questions anyone might have. It's a bit of a strange problem domain so might be hard to orient to what's even going on here if you're not already somewhat familiar. I have to step away for a little while now, but I'll check in later and try to address any questions/comments that emerge in the meantime.
- ecesena 4y agoAny reason to build from scratch vs contributing to arkworks?
- porcuquine 4y agoArkworks isn't really addressing the core of what Lurk provides. In theory, we could use Arkworks to implement a backend — but we are targeting Nova (https://github.com/microsoft/Nova https://github.com/microsoft/Nova), and I don't think Arkworks supports Nova currently. So the part we are building from scratch (the language itself) is at a higher level of abstraction. We like Nova's characteristics and are actively helping with aspects of its implementation so we can use it as soon as possible.
- ecesena 4y agoGreat, thank you for the details!
- noman-land 4y agoI've been following this space with extreme fascination from the perspective of a software engineer with no formal mathematics study so please take my questions with a grain of salt. My understanding is that a SNARK circuit is used to prove a series of computations without revealing any of the inputs or outputs. For example some people submit their salaries in encrypted form through a compiled zk circuit and receive an encrypted answer on the other side that that tells them which person earns the most money without revealing who earns what. Is my understanding correct that lurk would be used instead of something like circom where circom can create simple arithmetic zk circuits, whereas lurk can create circuits that can describe arbitrary computations? > Lurk constructs compound data using SNARK-friendly Poseidon hashes (provided by Neptune), so its data is naturally content-addressable. So lurk programs can point to previously proven (and published?) hashes to make provable claims about their calculations? Is this like "merkle-izing" computation? Sorry I'm just really trying to wrap my head around this.
- yellowapple 4y agoIn case I ain't alone in wondering what in tarnation a ZK-SNARK is: https://en.wikipedia.org/wiki/Non-interactive_zero-knowledge_proof https://en.wikipedia.org/wiki/Non-interactive_zero-knowledge...
- galfarragem 4y agoHow does this compare to pact[0]? [0] https://github.com/kadena-io/pact https://github.com/kadena-io/pact
- porcuquine 4y agoI'm not otherwise familiar with it, but the landing page linked describes Pact as 'a Turing-incomplete smart contract language'. Lurk could be used to implement smart contracts (with some glue connecting it to a blockchain — as we plan to do for the FVM), and we believe its characteristics will allow interesting applications in that domain. However, it's not inherently a 'smart contract language'. It looks like Pact considers Turing-incompleteness a benefit, since it makes it easier to reason about the cost of programs in a blockchain context. I don't see mention of zero-knowledge or succinct proofs, which are some of the more 'interesting' features of Lurk. For example, in a blockchain context, costs associated with program running time are less relevant for Lurk programs executed on chain because verification costs aren't affected by the running time of the program. All proofs created by a concrete instance of the language are verifiable in constant time. This is an example of how Turing-completeness and succinct proofs provide complementary benefits. It looks like Pact has a Lispy syntax and immutable data, so they seem to have that in common.
- sshine 4y agoWhat’s a zk-SNARK without a trusted setup? I thought zk-STARKs addressed this, but it seems like you’re continuing with a circuit model.
- porcuquine 4y agoNova (for example) doesn't require a trusted setup. The circuit is just a schematic description of the underlying computation. In the case of the Lurk core language, this computation is 'one reduction step of a Lurk evaluation' (https://github.com/lurk-lang/lurk-rs/blob/master/spec/reduction-notes.md https://github.com/lurk-lang/lurk-rs/blob/master/spec/reduct...). Coming up with a 'fixed computation' that yields general computation is part of the design problem for Lurk (or any other Lurk-like language). Even if we did need a per-circuit trusted setup (which we don't), we could perform such a setup for our core circuit and use it to prove arbitrary programs. For example, although we have not actually performed the trusted setup, we do have an example using Groth16 (which does require a trusted setup to be secure) and aggregates the potentially many discrete reduction steps to produce a succinct proof.
- emmorais 4y agoIndeed zk-STARK addresses it, but it is not the only alternative (Bulletproofs, Nova, Halo2 also doesn't require a trusted setup). All zk-SNARKs need some kind of intermediate representation that looks like a circuit, then this characteristic has nothing to do with the trusted setup requirement.
- 0xcb0 4y agoJust wow! This sounds very interesting and I can think of lots of applications this could be useful in. And even the facts that it is in LISP makes it so cool! Thanks @porcuquine for the hard work you are putting in here!
- osense 4y agoCan someone provide a good real-world usecase for zero-knowledge proofs? I struggle a bit with grasping the usefulness of this concept.
- b_fiive 4y agozero-knowledge proofs are a means of showing you know a thing without exposing the thing itself. In the context of lurk, the "thing" is executed computation. That's useful because sometimes computation is expensive, and we'd like to run that expensive computation, anywhere, once, and be able to re-use the result.
- osense 4y agoThank you!
- masukomi 4y agofor those of you who, like me, had no clue what a ZK-SNARK was/is > Zk-SNARK is an acronym that stands for “Zero-Knowledge Succinct Non-Interactive Argument of Knowledge.” A zk-SNARK is a cryptographic proof that allows one party to prove it possesses certain information without revealing that information. This proof is made possible using a secret key created before the transaction takes place. It is used as part of the protocol for the cryptocurrency, Zcash. more details here: https://www.investopedia.com/terms/z/zksnark.asp https://www.investopedia.com/terms/z/zksnark.asp