Show HN: Formal Verification for Machine Learning Models Using Lean 4

Formal Verification of Machine Learning Models in Lean Welcome to the Formal Verification of Machine Learning Models in Lean project. This repository provides a framework for specifying and proving properties—such as robustness, fairness, and interpretability—of machine learning models using Lean 4. A live, interactive webpage is available at: proof-pipeline-interactor.lovable.app Overview In high-stakes applications (e.g., healthcare, finance, autonomous systems), ensuring that machine learning models meet strict reliability and fairness properties is essential. This project provides: Lean Library : Formal definitions for neural networks, linear models, decision trees, and advanced models (ConvNets, RNNs, Transformers), along with properties like adversarial robustness, fairness, interpretability, monotonicity, and sensitivity analysis. : Formal definitions for neural networks, linear models, decision trees, and advanced models (ConvNets, RNNs, Transformers), along with properti ... Read full article.