Home > News > China News > Robots also learn the second l.....
Contact us
CHINA TOPWIN INDUSTRY CO.,LTD. (ChinaGeterTechnologyCo.,LTD)was established in 2002. In the attitude of "Study for our innovation and advancement. Pro...Contact Now


Robots also learn the second language, which helps improve natural language skills

  • Author:chinatopwin
  • Source:chinatopwin
  • Release on:2017-08-10
From Alexa and Siri to countless chat robots and automated customer support systems, 
computers are learning to speak. The only problem is that computers are often confused.

A research team at Salesforce has come up with a clever way to improve the performance 
of many modern language programs, namely, teach another language before training the 
algorithm to accomplish other tasks.
Keeping coherent conversations in church machines remains a major challenge in the field 
of artificial intelligence, because it is often necessary to have a broader understanding or 
common sense of the world in order to understand clearly the meaning of spoken or written 
In fact, the training of machine learning system for transformation between two languages,
 it can automatically teach collocation relationship and appropriate context. When this system 
is used as a training to dialogue, emotional or contains text in the machine learning based 
detection system, its performance is far better than the an ab initio training system.
"We're using Machine Translation data, and we're basically teaching the model how to 
understand vocabulary and context," said Richard Sugar, chief scientist at Salesforce and an 
expert on machine learning and language."
This work is machine learning progress an example of artificial intelligence system of language 
skills. Many computer vision system using deep learning network based on some form of pre 
training, according to Su Ge said, Machine Translation may provide a similar way to guide the 
natural language system.
Salesforce is a sales, management of customer interaction between marketing and business 
online platform, through the platform of Einstein has provided a series of artificial intelligence 
tools. One is the automatic classification of email or chat messages emotional tools, another 
is the staff according to his or her previous activities in pursuit of potential customers the 

He believes the discovery will help improve the natural language capabilities of the Einstein 
platform. "It's very useful for chat bots and automated customer support," he says"
Salesforce researchers trained a translation between English and German deep learning 
system. It involves a large number of translation input document into a multi-layer neural 
network, and continuously adjust the network parameters, until it has learned to automatically 
generate a decent translation. This system is represented by a vector of words. This is a 
common way of encoding and parsing the meaning of the text.
Then, the researchers conducted training on the bilingual network, let them do all kinds of 
things: a text emotion; to classify different types of problems; and then answer the questions.
 They also found that pre training network more than they did not learn the second language 
Machine Translation data set is particularly large, which contributes to machine learning this 
challenge. The researchers involved in the project Salesforce Mccain said, "there is an 
important connection between translation and other language translation. The data set is very 
common, they contain information that is very useful for Natural Language Processing."