Renegade
A K-nearest neighbors machine learning algorithm with a learned metric, designed to be parameter-free and data-efficient for use cases with limited training data
As defined in Freenet - Services Without Servers
Where it comes up · 1
Most important claims · 4 of 4Search all claims for “Renegade” →
- Freenet solves the contract discovery routing problem by treating it as a supervised learning problem using the Renegade algorithm, a K-nearest neighbors algorithm with a learned metric implemented in Rust.Freenet - Services Without Servers
- If an undersea cable degrades or goes down, Renegade will quickly detect that previously fast peers are now slower and adapt Freenet peer routing very quickly.Freenet - Services Without Servers
- Renegade is very data efficient compared to neural networks which require vast amounts of data to train, making Renegade suitable for Freenet's decentralized learning context.Freenet - Services Without Servers
- Renegade is available under the LGPL license, is implemented in Rust, and was designed to perform well with default settings without needing multivariate testing or tuning.Freenet - Services Without Servers