I Know What You Trained Last Summer: A Survey on Stealing Machine Learning Models and Defences. (arXiv:2206.08451v1 [cs.LG])

Machine Learning-as-a-Service (MLaaS) has become a widespread paradigm,
making even the most complex machine learning models available for clients via
e.g. a pay-per-query principle. This allows users to avoid time-consuming
processes of data collection, hyperparameter tuning, and model training.
However, by giving their customers access to the (predictions of their) models,
MLaaS providers endanger their intellectual property, such as sensitive
training data, optimised hyperparameters, or learned model parameters.
Adversaries can create a copy of the model with (almost) identical behavior …

lg machine machine learning machine learning models stealing summer survey