公开(公告)号：WO2021119074A1

公开(公告)日：2021-06-17

申请号：PCT/US2020/063926

申请日：2020-12-09

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： MIN, Renqiang ,  MALON, Christopher ,  PENGYU, Cheng

IPC: G06N3/08 ,  G06N20/00

Abstract: A computer-implemented method is provided for disentangled data generation. The method includes accessing (310), by a bidirectional Long Short-Term Memory (LSTM) with a multi-head attention mechanism, a dataset including a plurality of pairs each formed from a given one of a plurality of input text structures and given one of a plurality of style labels for the plurality of input text structures. The method further includes training (320) the bidirectional LSTM as an encoder to disentangle a sequential text input into disentangled representations comprising a content embedding and a style embedding based on a subset of the dataset. The method also includes training (350) a unidirectional LSTM as a decoder to generate a next text structure prediction for the sequential text input based on previously generated text structure information and a current word, from a disentangled representation with the content embedding and the style embedding.