, overfitting, and vulnerability to adversarial perturbations - in order to create a method for the secure and efficient training of remotely deployed neural networks over unsecured channels.  ...  We leverage what are typically considered the worst qualities of deep learning algorithms - high computational cost, requirement for large data, no explainability, high dependence on hyper-parameter choice  ...  Conclusion and future work We have proposed a method for producing synthetic data that can be used to securely and efficiently train remotely deployed neural networks over unsecured channels.  ...