Abstract: The function input means 71 accepts input of a cost function that calculates a cost incurred in at least one of a selection of a candidate relay point and a design of a route, the cost function being represented as a linear sum of terms weighted by degree of importance attached to each of the features that an expert is assumed to intend when selecting the candidate relay point and design of the route. The learning means 72 learns the cost function by inverse reinforcement learning using training data that includes relay point information which is data that maps information indicating a relay point with surrounding information of the relay point and usage information of the relay point, and route result information which is result data of a route that pass through each relay point.

Abstract: An input unit 81 receives inputs of pre-change performance data acquired by a device before a change and post-change performance data acquired by the device after having undergone the change, through control using a first cost function. An update unit 82 generates a second cost function obtained by updating the first cost function in such a way as to reduce a difference between the pre-change performance data and the post-change performance data. In the process, the update unit 82 generates the second cost function obtained by updating the first cost function by estimating an error that occurs in an output value of the device included in the first cost function before and after the change to the device.

Abstract: A compound selected from the group consisting of compounds represented by the formulas below or pharmaceutically acceptable salt thereof is useful in PET imaging for AMPA receptor.