Yearn Machine Learning: Bridging the Gap Between A.I. and DeFi

Yield farming has become a lucrative use case for cryptocurrency holders which allows them to earn concurrent interest on their otherwise idle tokens/coins. But this doesn’t come without risk. The higher the APY (annual percentage yield) for these farming and staking pools, the higher the risk exposure the holder faces.

With this risk vs. benefit problem being largely ignored, a new project called Yearn Machine Learning (YML) has taken on the job of protecting yield farmers by using specific parameters and multi-agent enforcement learning. The developers of YML are using historical data and algorithms to help create models that have the capabilities to predict which yield pools have the most optimization.

Image for post
Image for post

Yearn Machine Learning (YML, Contract Address: 0xD7E48d68DCE805C97bB66B2AfFaDcdFcd47E1F95) is an ERC20 token bringing yield farmers a new way to earn the most APY. The project is utilizing its own algorithms built by the developers that will help point holders in the right direction, meaning providing data and a solution using said data to receive the highest APY via the most efficient pools available. The developers call this process of using machine learning “Multi-agent Reinforcement Learning". One of the algorithms used to process the data needed to weigh risk vs. benefit is the Deterministic Policy Gradient (DDPG).

This algorithm simultaneously learns a Q-Function and the policy in question, in this instance the code surrounding yield farming protocols, in order for the Q-Function to analyze the policy in question.

These Deep Learning Reinforcement calculations create a form of artificial intelligence that can help predict which pools will provide the highest return percentage for yield farmers. The AI developed by the team at YML utilizes multiple agents that perceive the yield opportunities with sensors written in the code and acts on these opportunities with effectors. YML’s code, written in C++, has recently been audited by Solidity.Finance and the full report can be reviewed here for your convenience.

All of the things YML does in the background is compacted and presented to the user in a polished dashboard which displays your ROI (return on investment rates will be subject to change as this rate is not a permanent one as a result of the agents finding the most profitable farming pools with the least amount of risk) along with other vital data so yield farmers can monitor their balances.

Image for post
Image for post

Yearn Machine Learning follows the distribution model outlined below:

  • Max Supply — 30,000 YML Tokens
  • Circulating Supply — 10,500 YML Tokens
  • Team Tokens — 1,500 YML
  • Marketing/Development — 3,000 YML Tokens
  • Presale Tokens — 3,000 YML Tokens
  • Uniswap Liquidity — 3,000 YML Tokens
  • Locked YML (Unicrypt) — 19,500 YML Tokens
Image for post
Image for post

Holders of YML are able to benefit off of the capabilities its machine learning provides while also enjoying staking rewards when contributing to the YML ecosystem. Holders who stake their tokens will be eligible to vote on governance policies and updated procedures, giving the project a dose decentralization. Voting will be conducted off chain to avoid gas-related issues that other DAO’s face by using Snapshot to keep voting transparent, fair, and efficient.

Staking rewards received for partaking in the network are distributed via a third party contract once a set amount has been accumulated through staking. The accumulated rewards will be distributed accordingly in correspondence with reward expectations per holder. The following APY schedule is provided for staking YML:

  • Daily — 0.11%
  • Monthly — 3.33%
  • Yearly — 40%

It is important to note that the YML team will receive a percentage of profits due to its decentralized structure as outlined in their whitepaper.

Providing a system which creates lower risk and increased opportunities at maximizing profit can be a game changer in the DeFi realm. DeFi has exploded and expanded into one of the biggest niches of cryptocurrency, but one of its biggest and most glaring problems is the risk involved. Lowering this risk is very difficult considering the decentralized nature of most yield farming and staking platforms.

From an ethical standpoint, solving this problem comes with decreased decentralization in most cases which goes against the spirit of cryptocurrency and DeFi specifically. YML solves this problem by using proven machine learning methods which make SeFi a safer and more sustainable place. More importantly it still provides a level of decentralization that makes token holders comfortable. Building platforms that help yield farmers through machine learning is one of the most innovative and complex ways I’ve come across. Artificial Intelligence is an incredibly useful tool, and I find it quite impressive that the developers at YML have created a way to take advantage of it.

Pertinent Links:

  • Website —
  • Twitter —
  • Telegram News —
  • Telegram Community —
  • Discord —
  • Etherscan —
  • YML Forum —

(I write articles, reviews, and Litepapers for legitimate, interesting, up and coming cryptocurrency projects. Feel free to PM me to review your project. Thank you!)

Disclaimer: This is not financial advice. The sole purpose of this post/article is to provide and create and informative and educated discussion regarding the project in question. Invest at your own risk.

Written by

Cryptocurrency Enthusiast

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store