MULTI-TASK LEARNING NETWORK AND METHOD BASED ON MULTI-TASK LEARNING NETWORK

Title:

MULTI-TASK LEARNING NETWORK AND METHOD BASED ON MULTI-TASK LEARNING NETWORK

Document Type and Number:

Japanese Patent JP2022121385

Kind Code:

A

Abstract:

To provide a multi-task learning network and a method based on the multi-task learning network.SOLUTION: The multi-task learning network includes a task network and a merging network. The task network includes a plurality of sub-task networks, and the plurality of sub-task networks generates corresponding feature vectors with respect to the same input, and the merging network merges the corresponding feature vectors and performs optimization processing on a feature vector resulting from merging to obtain a final output. Each of the sub-task networks includes a coding layer and a task layer, and the coding layer converts an input to a coded vector including dimensionality information and transmits the coded vector to the task layer, and the task layer performs processing on the coded vector to generate a feature vector.SELECTED DRAWING: Figure 3

Inventors:

ZHENG ZHONGGUANG
FANG LU
CHAO ILYN
SON SHUN

Application Number:

JP2022012018A

Publication Date:

August 19, 2022

Filing Date:

January 28, 2022

Export Citation:

Click for automatic bibliography generation Help

Assignee:

FUJITSU LTD

International Classes:

G06N3/04; G06N3/08

Attorney, Agent or Firm:

Tadashige Ito
Tadahiko Ito

Previous Patent: Electric vehicle charging method, system, and storage medium

Next Patent: Speaker Diarization Correction Method and System Utilizing Text-Based Speaker Change Detection