Neural Net AI

VyFinance plans to bring Auto-Harvesting to Cardano, which will be made as efficient as possible thanks to our Neural Net. The Neural Net will be used to measure risk across yield-farms, using impermanent loss, and a self-improving algorithm, to find the best Yield Farm’s to Auto-Harvest! The Vy-Ecosystem This consists of 4 components, all linked through smart contracts. The Auto-Harvester, The DEX (decentralized exchange), The Bar and LMS (Learning Management System). The Auto-Harvester (AH) is a smart contract that will manage yield farming/liquidity pool investing across multiple platforms. Users will deposit their Cardano into our AH, and will be given access to our dashboard. This will contain all the information regarding their deposit, current portfolio value, and pending harvest. The auto-harvester distributes rewards once weekly, and 15% of all earnings are shared to the VyFi Bar.
The Auto-Harvester exists to provide two functions: Assist new-comers in accessing more complex tools; and to give an option for traders to automatically generate the average return offered across the ADA ecosystem. The focus will be on simplicity and applicability for newcomers, as our principal focus is assisting on-ramp new users to the DeFi ecosystem, and guide them safely through the process of engaging with DeFi as a whole. Auto-Harvesters are automated to a point where fine-tuning isn't necessary, thus, there isn't a need for an "advanced mode." - the Neural Net will be constantly performing fine-tuning and then verifying the tuning. VyFinance is currently working with the minds behind Propella.ai to produce a Neural Net AI. This AI will be used to both help International Markets calculate risk - as well as for the Auto-Harvester. The Neural Net will be using ARIMA, and other proprietary methods, to find the best liquidity pools, and auto-stake to it. This is reminiscent of other liquidity aggregators, but with a more advanced AI technology managing the system.
Down the line, the Neural Net will also provide functionality to the other trading platforms. Mathematics, such as finding and displaying impermanent loss, can be made more efficient with this design. In general, there will be metrics that have been produced from munged data that can be oracalised, and used in trading strategies or to aid in the analysis of the crypto markets. The neural net AI will take some time to develop, as it needs to be taught. Users should expect to see more around mid-2023.