Applying AI to blockchain as a POC in response to https://www.youtube.com/watch?v=5Tr13l0O1Ws
Switch branches/tags
Nothing to show
Clone or download
Pull request Compare This branch is 1 commit ahead of DecentricCorp:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
app
assets
contracts
media
scenarios
views
.gitignore
Dockerfile
Readme.md
Readme.old.md
click.py
config.js
dynamic-api.js
entry.sh
files.js
index.js
package-lock.json
package.json
requirements.txt

Readme.md

Circuits of Value (http://cov.al) + Unspecified (http://unspecified.life)

Blockchain and Machine Learning Collaboration

Known Vulnerabilities


Unspecified LLC:

Applying Machine learning and Ai towards the optimization of software delivery.

Circuits of Value:

Framework for a new internet that’s distributed, decentralized and trustless. Turning traditional crypto currency models inside out and bridging the decide between utility and revolutionary.


Machine Learning technologies

  • Python
  • Gym
  • Universe
  • World of Bits

Blockchain technologies

  • NodeJS
  • Ethereum
  • Test RPC
  • Coval-Solidity

Executive Summary

Using Test RPC We create a light weight ethereum emulator to deploy smart contracts into.

We next deploy the ClickButton Contract using the Coval-Solidity Deployment CLI Tool

We then start the REST server from the CLI allowing interaction with the deployed contracts.

Next (having previously set up a working gym environment we run click.py)

Click.py loads the wob.mini.ClickButton-v0 environment to begin training a model to be able to click the button requested.

Each iteration of a task records the PointerEvent's within ethereum storage for later retrieval and and historical purposes.

Click.py communicates with the contract via REST over http here: http://127.0.0.1:4333/v1/MachineLearning/ClickButton/SaveEvent/exec?args=roundId,eventId,x,y,mask


How to run locally

  • Install Ethereum testrpc in one terminal npm install -g ethereumjs-testrpc
  • Run testrpc (terminal 1) testrpc

  • Install Coval Solidity npm install
  • Deploy scenario "MachineLearning" (terminal 2) npm start

- Launch REST server from tools menu after deploying contract - Alternatively run from terminal `npm run web`
  • Install gym and universe pip install gym universe

  • Launch click.py (terminal 3) python click.py

How to run within Docker