https://github.com/seiyakobayashi/zkml-optimization

Optimization for on-chain private machine learning.
https://github.com/seiyakobayashi/zkml-optimization

blockchain ethereum machine-learning merkle-proof merkle-tree neural-network optimization-algorithms zk zk-snarks zkml zkp

Last synced: 3 months ago
JSON representation

Optimization for on-chain private machine learning.

• Host: GitHub
• URL: https://github.com/seiyakobayashi/zkml-optimization
• Owner: SeiyaKobayashi
• License: mit
• Created: 2022-11-05T01:38:25.000Z (about 2 years ago)
• Default Branch: develop
• Last Pushed: 2023-10-26T05:05:51.000Z (over 1 year ago)
• Last Synced: 2024-10-11T08:24:19.439Z (4 months ago)
• Topics: blockchain, ethereum, machine-learning, merkle-proof, merkle-tree, neural-network, optimization-algorithms, zk, zk-snarks, zkml, zkp
• Language: TypeScript
• Homepage:
• Size: 2.07 MB
• Stars: 13
• Watchers: 5
• Forks: 1
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# zkml-optimization

Optimization of on-chain private machine learning (public model × private data).

## Architecture Diagrams

### Architecture Diagram of Verifier

![Verifier](./docs/v1/arch-diagram-verifier.png "verifier")

### Architecture Diagram of Prover

![Prover](./docs/v1/arch-diagram-prover.png "prover")

### Architecture Diagram of Model Developer

![Model Developer](./docs/v1/arch-diagram-model-developer.png "model developer")

## Prerequisites

All commands in this `README` (except this section) are assumed to be called from the root directory.

```sh

# cd into this repo

$ cd zkml-optimization

```

## How to Use Demo CLI App

### 0. Installation

Install all the dependencies using `yarn`.

```sh

$ yarn

```

### 1. Steps for Model Developers

#### 1-A. Demo Model & Circuit

We have a demo ML model called `demo.h5` in `./models` directory. This is a classical classification model for MNIST dataset, and is a slightly modified version of [this model](https://github.com/socathie/keras2circom/blob/main/models/best_practice.h5). If needed, you can take a look at its implementation from `demo.ipynb` in `./models` directory.

We are using [keras2circom](https://github.com/socathie/keras2circom) as a way to convert model files (`.h5`) into circom-compatible circuits. Follow the steps in keras2circom's [README](https://github.com/socathie/keras2circom#keras2circom) to convert models into circuits. The demo circuit files auto-generated from the demo model are `demo-circuit.circom` and `demo-circuit-json` in `./circuits/` directory.

We need a ptau file for our trusted-setup. Clone `powersOfTau28_hez_final_17.ptau` from [this repo](https://github.com/iden3/snarkjs#7-prepare-phase-2), and place it in `./circuits` directory.

#### 1-B. Circuit Compilation, ZKey Generation, IPFS, Model Registration

Go to `./model-developers` directory, and follow the instructions in `README.md` for compiling demo circuit, generating zkeys from them, uploading files to IPFS, and registering the resulting hash to our custom verifier contract.

### 2. Steps for Prover Clients

Go to `./provers` directory, and follow the instructions in `README.md` for encoding testing results as a Merkle tree, generating Merkle proofs and ZKPs given the random challenge, and sending proofs to our custom verifier contract for revealing/verification.

## Contributions

### Adding Packages

To minimize side-effects of changes to packages, make sure to pin exact versions when adding packages.

```sh

# example: add package 'xxx' to root project as a dev dependency 

$ yarn add -W -D -E xxx

```