Abstract: A learning device includes a classification unit that classifies data to be determined by using a learner configured to classify data based on a predetermined feature among features included in the data. The learning device includes an estimation unit that estimates, from classification results according to the classification unit, the behavior of the probability of a mistake occurring in the classification results according to the classification unit based on a large deviation principle. The learning device includes a determination unit that determines, based on the behavior estimated by the estimation unit, whether to add a new feature to an object to be learned for the learner.

Abstract: According to one aspect of an embodiment an encryption device includes an acquisition unit that acquire information that is to be an encryption object. The encryption device includes an encryption unit that apply a plurality of functions to the acquired information in order of having a quasi-periodic structure to encrypt the information.

Abstract: A learning device includes a classification unit that classifies data to be determined by using a learner configured to classify data based on a predetermined feature among features included in the data. The learning device includes an estimation unit that estimates, from classification results according to the classification unit, the behavior of the probability of a mistake occurring in the classification results according to the classification unit based on a large deviation principle. The learning device includes a determination unit that determines, based on the behavior estimated by the estimation unit, whether to add a new feature to an object to be learned for the learner.