LEARNING APPARATUS AND METHOD FOR BIDIRECTIONAL LEARNING OF PREDICTIVE MODEL BASED ON DATA SEQUENCE

摘要

To accurately learn a predictive model based on a data sequence, provided is a method comprising generating, based on a first model for performing a learning process by sequentially inputting each piece of input data from a first input data sequence in which are arranged pieces of input data including a plurality of input values, a second model for learning a first learning target parameter included in the first model by inputting, in an order differing from an order in the first model, each piece of input data from a second input data sequence in which are arranged pieces of input data including a plurality of input values; performing a learning process using both the first model and the second model; and outputting the first model that has been learned. Also provided are a computer program product and apparatus.