💸 Algovera: Machine learning for predicting the borrowing cost across different DeFi protocols
Is it possible to use machine learning to predict the borrowing cost across different DeFi protocols? It would be useful to be able to present the best borrowing and lending opportunities in DeFi, as well as show predictions of how that market will behave in the future. The main goal of this strategy is to capture the yield given out by the protocol without taking exposure to liquidation risk or too much market exposure. Any processed datasets, results/analysis and algorithms developed will be published on Ocean marketplace, with the hope of helping DeFi investors make better decisions using quality information. The scope of the project can be found here.
Contents
👪 Community
Hacking Sessions
- Algovera hacking sessions: Wednesdays at 17:00 UTC. 60 min. Zoom link, Meeting ID 834 978 0070, Passcode 370768. Anyone is welcome to drop in!
👋 - Catch up on previous sessions in the Hacking Sessions playlist of our YouTube
- Join the discussion in the #
🖥 -hacking-sessions channel of our Discord
Algovera
Algovera is a community of individuals working to facilitate and accelerate the development of decentralised AI products and research.
Website | Notion | Discord | Calendar | Twitter | YouTube
🏗 Initial Setup
Set up environment
Open a new terminal and:
#clone repo
git clone https://github.com/AlgoveraAI/DeFi.git
cd DeFi-borrowing-cost-prediction🤖 Resources
- Successful Algovera Grants proposal1, proposal2
- Project scope (thanks Iago!)
- Video overview of machine learning techniques that can be used to classify/cluster user accounts as part of the DeFi & Cross-chain Interoperability Hackathon on Gitcoin
- Good summary for high level options to query historical blockchain data in the Synthetix docs
- Incredibly informative article describing statistical approach to choosing optimal Uni V3 LP positions
- Good resource for labeled blockchain data in the Flipside docs
🏛 License
The license is MIT. Details