Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates

Open in new window