Inference apparatus for performing inference with a machine learning model, inference method and computer-readable storage medium

An inference method comprising;

• inputting input data to a first machine learning model trained using training data;
• performing inference by the first machine learning model to output inference data indicating a result of the inference;
• determining, using a second machine learning model, whether the input data is the training data or not based on the inference data;
• and
• outputting, by the second machine learning model, the inference data as output data based on determining that the input data is not the training data, and outputting an alternative data different from the inference data as the output data based on determining that the input data is the training data
• wherein the inference data is an n-dimensional score vector including n scores, n being an integer greater than or equal to 2,
• the alternative data is an n-dimensional score vector having the same number of dimensions as that of the inference data, and
• an order of a component of a top score of the inference data is matched with that of the alternative data;
• and
• wherein components of the alternative data are n random numbers, and
• the n random numbers are rearranged in accordance with a relationship between a magnitude of the n scores of the inference data such that the order of the magnitudes of the n scores of the inference data is the same as that of the alternative data.

A non-transitory computer-readable storage medium storing a program that causes a computer to execute an inference method:

• the method comprising;
• inputting input data to a first machine learning model trained using training data;
• performing inference by the first machine learning model to output inference data indicating a result of the inference;
• determining, using a second machine learning model, whether the input data is the training data or not based on the inference data;
• and
• outputting, by the second machine learning model, the inference data as output data based on determining that the input data is not the training data, and an alternative data different from the inference data as the output data based on determining that the input data is the training data
• wherein the inference data is an n-dimensional score vector including n scores, n being an integer greater than or equal to 2,
• the alternative data is an n-dimensional score vector having the same number of dimensions as that of the inference data, and
• an order of a component of a top score of the inference data is matched with that of the alternative data;
• and
• wherein components of the alternative data are n random numbers, and
• the n random numbers are rearranged in accordance with a relationship between a magnitude of the n scores of the inference data such that the order of the magnitudes of the n scores of the inference data is the same as that of the alternative data.