Invention Grant
- Patent Title: Corrected mean-covariance RBMs and general high-order semi-RBMs for large-scale collaborative filtering and prediction
-
Application No.: US15088312Application Date: 2016-04-01
-
Publication No.: US10339442B2Publication Date: 2019-07-02
- Inventor: Renqiang Min , Eric Cosatto
- Applicant: NEC Laboratories America, Inc.
- Applicant Address: JP
- Assignee: NEC Corporation
- Current Assignee: NEC Corporation
- Current Assignee Address: JP
- Agent Joseph Kolodka
- Main IPC: G06N3/04
- IPC: G06N3/04
Abstract:
Systems and methods are disclosed for operating a Restricted Boltzmann Machine (RBM) by determining a corrected energy function of high-order semi-RBMs (hs-RBMs) without self-interaction; performing distributed pre-training of the hs-RBM; adjusting weights of the hs-RBM using contrastive divergence; generating predictions by Gibbs Sampling or by determining conditional probabilities with hidden units integrated out; and generating predictions.
Public/Granted literature
Information query
