The PME trains a machine learning model M that infers encrypted prices from associated features. This model can be applied locally to each user. By capturing the context of encrypted prices, PME can run more probed ad campaigns and determine which features are most important. These parameters can then be used to model encrypted prices.