Apparatus and method for learning a model corresponding to time-series moving image/video input data

摘要

A dynamic time-evolution Boltzmann machine capable of learning is provided. Aspects include acquiring a time-series input data and supplying a plurality of input values of input data of the time-series input data at one time point to a plurality of nodes of the mode. Aspects also include computing, based on an input data sequence before the one time point in the time-series input data and a weight parameter between each of a plurality of input values of input data of the input data sequence and a corresponding one of the plurality of nodes of the model, a conditional probability of the input value at the one time point given that the input data sequence has occurred. Aspects further include adjusting the weight parameter so as to increase a conditional probability of occurrence of the input data at the one time point given that the input data sequence has occurred.

主权项

1. A computer implemented learning method for learning a model corresponding to time-series input data, comprising:
acquiring, by a processor coupled to a memory, the time-series input data, wherein the time series input data is moving image data; supplying, as training data, a plurality of input values at one time point to a plurality of nodes of the model corresponding to input data of the time series data at the one time point; computing, via the processor, a conditional probability of an occurrence of an input data sequence for each of the plurality of input values at the one time point;
wherein the computing the conditional probability is based on an input data sequence before the one time point in the time series input data;wherein the computing the conditional probability is based on a weight parameter between each of a plurality of input values and a corresponding one of the plurality of nodes of the model; adjusting the weight parameter so as to increase the conditional probability of the occurrence of the input data sequence for each of the plurality of input values at the one time point; and predicting a next occurrence of the input data sequence for each of the plurality of input values at a next one time point.