Invention Grant
- Patent Title: Staged training of neural networks for improved time series prediction performance
-
Application No.: US16727023Application Date: 2019-12-26
-
Publication No.: US10740395B2Publication Date: 2020-08-11
- Inventor: Henry Gabriel Victor Bequet , Jacques Rioux , John Alejandro Izquierdo , Huina Chen , Juan Du
- Applicant: SAS Institute Inc.
- Applicant Address: US NC Cary
- Assignee: SAS INSTITUTE INC.
- Current Assignee: SAS INSTITUTE INC.
- Current Assignee Address: US NC Cary
- Main IPC: G06F9/46
- IPC: G06F9/46 ; G06F16/901 ; H04L29/08 ; G06F16/903

Abstract:
An apparatus includes a processor to: train a first neural network of a chain to generate first configuration data including first trained parameters, wherein the chain performs an analytical function generating a set of output values from a set of input values, each neural network has inputs to receive the set of input values and outputs to output a portion of the set of output values, and the neural networks are ordered from the first at the head to a last neural network at the tail, and are interconnected so that each neural network additionally receives the outputs of a preceding neural network; train, using the first configuration data, a next neural network in the chain ordering to generate next configuration data including next trained parameters; and use at least the first and next configuration data and data indicating the interconnections to instantiate the chain to perform the analytical function.
Public/Granted literature
- US20200133977A1 STAGED TRAINING OF NEURAL NETWORKS FOR IMPROVED TIME SERIES PREDICTION PERFORMANCE Public/Granted day:2020-04-30
Information query