Unlock instant, AI-driven research and patent intelligence for your innovation.

Knowledge-driven and self-supervising system for questions-answers

A technology of questioning and global knowledge, applied in the field of machine learning systems, can solve problems such as the inability of neural language models to predict

Pending Publication Date: 2022-05-06
ROBERT BOSCH GMBH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As such, these neural language models may not be able to make reliable predictions on other question-answering tasks that may differ in format, type of knowledge, or both

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Knowledge-driven and self-supervising system for questions-answers
  • Knowledge-driven and self-supervising system for questions-answers
  • Knowledge-driven and self-supervising system for questions-answers

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] The embodiments described herein, having been shown and described by way of example, and their many advantages will be understood from the foregoing description, and it will be apparent that without departing from the disclosed subject matter or sacrificing one or more of its advantages, Here, various changes may be made in the form, construction and arrangement of components. Rather, the form of description of these embodiments is illustrative only. These embodiments are susceptible to various modifications and alternative forms, and the following claims are intended to embrace and embrace such changes, and are not limited to the particular forms disclosed, but cover all that fall within the spirit and scope of the disclosure. Modifications, Equivalents and Alternatives.

[0016] figure 1 is an illustration of an example of a system 100 with a neural-symbolic framework 200 for query tasks, according to an example embodiment. System 100 is configured to pre-train (or...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a knowledge-driven and self-supervising system for questions-answers. A computer-implemented system and method relate to natural language processing. Computer-implemented systems and methods are configured to obtain a current data structure from a global knowledge graph that includes various knowledge graphs. The current data structure includes a current header element, a current relationship element, and a current tail element. A sentence is obtained based on the current data structure. A question is generated by removing a current tail element from the sentence. A correct answer is generated for the question. The correct answer comprises the current tail element. A pool of data structures is extracted from the global knowledge graph based on the interferent standard set. The interferent standard set ensures that each extracted data structure includes a current relational element. Tail elements are extracted from the pool of data structures to create a pool of interferent candidates. A set of interferent is selected from the pool of interferent candidates. A query task is created, the query task including questions and a response option set. The set of response options includes correct answers and a set of interferents.

Description

technical field [0001] The present disclosure relates generally to machine learning systems, and more particularly to machine learning systems configured for natural language processing. Background technique [0002] In general, there are some machine learning systems that include neural language models configured for question-answering tasks. However, some worry that these neural language models are overfit to specific question-answering tasks where they are trained without learning to exploit external knowledge and / or perform general semantic reasoning. More specifically, these neural language models tend to overfit to the specific question-answer format of the training data and / or the specific knowledge type of the training data. As such, these neural language models may not be able to make reliable predictions for other question-answering tasks that may differ in format, type of knowledge, or both. Contents of the invention [0003] The following is an overview of ce...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/33G06F16/332G06N3/08G06N20/00
CPCG06F16/3329G06F16/3344G06N20/00G06N3/08G06N3/088G06N5/022G06N3/042G06F16/9536G06N5/02
Inventor A·奥尔特拉马里J·弗朗西斯K·马F·伊利耶夫斯基
Owner ROBERT BOSCH GMBH