Event occurrence time learning device, event occurrence time estimation device, event occurrence time estimation method, event occurrence time learning program, and event occurrence time estimation program

摘要

PROBLEM TO BE SOLVED: To estimate an event occurrence from time series information which is high-dimensional information such as an image. A hazard estimation unit 11 includes a time-series image group in which an event occurrence time is given in advance, a time-series image group in which an event does not occur, and a time-series image group in which an event occurs. The likelihood of event occurrence for each of the plurality of time series image groups is estimated according to a hazard function, and the parameter estimation unit 12 determines the occurrence time of the event given for each of the plurality of time series image groups and a plurality of time series image groups. The parameters of the hazard function are estimated so as to optimize the likelihood function represented by including the estimated event likelihood for each of the time series images. [Selection diagram] Fig. 1