Skip to content

๐ŸŒ  ETH Brussels Contest : Decentralized Machine Learning Protocol adapted to both terrestrial and spatial context to enable the collaboration of different military forces.

License

Notifications You must be signed in to change notification settings

Ophiase/Astral-ZK-ML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

79 Commits
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

Astral-ZK-ML

Sponsorized by ETH Global - Brussels.
Prize : Nethermind - Best zkML project

Decentralized Machine Learning Protocol adapted to both terrestrial and spatial context.
The models and the two smartcontracts utilizing them are written in the ZK-Proovable language Cairo.
In the future, the smart contracts will be deployed on specific L3 for each planets/asteroids using Madara.

Introduction

  • โœ… ZK-ML tools : Built in Cairo 2.6.4 for contracts on starknet-compatible blockchain
    • โœ… Implemented on blockchain
      • Basic Algebra + Linear Algebra
        • Principal types: Signed Integer, Float (WFloat), Vector, Matrix
        • Numerous High level operations to manipulate those types.
        • Uniform pseudo random generator (to initialize weights)
          • TODO: add more distributions (normal, poisson, $\dots$)
      • Multi Layer Perceptron (ie. regular Neural Network)
        • eg: Star type identification
    • โŒ Not implemented in cairo yet
      • Convolutional Neural Network
        • for Crater identification
      • Graph Attention Network
        • for Constellation identification
  • โŒ Additional proof of concept for tasks assigned to spatial decentralized robots
    • Crater identification (Computer Vision)
    • Constellation identification (Computer Vision)
    • Microorganism analysis (Computer Vision)
  • Decentralized Smart Contracts : Variants of the smart contract :
    • โŒ Fully Decentralized (not implemented yet)
      • Based on financial incentive (ie. egoistic incentive)
      • Terestrial variant adapted to Starknet L2.
    • โŒ Country Wise Decentralized (not implemented yet)
      • Assumed resistance to Byzantine fault
      • Spacial variant adapted to a specific L3 for a planet/satellite owned by multiples countries.
  • Demonstration Client
    • โŒ Not implemented yet

Documentation

About

๐ŸŒ  ETH Brussels Contest : Decentralized Machine Learning Protocol adapted to both terrestrial and spatial context to enable the collaboration of different military forces.

Topics

Resources

License

Stars

Watchers

Forks