Recently, the 2016 Baidu Yunzhi Summit (ABC Summit) was officially held at the Beijing International Convention Center. The summit focused on the three major industries of cloud computing, artificial intelligence and big data with the theme of “intelligence and unlimited possibilitiesâ€. Liu Wei, co-director of Baidu Cloud Business Department, delivered a speech titled “ABC Times†and highlighted Baidu Cloud Artificial Intelligence platform-level solution “Tianzhi†to fully open Baidu’s brain capabilities.
Tianzhi is the fourth platform-level solution released by Baidu Cloud after the three intelligent platforms “Tianjiâ€, “Tianxiang†and “Tiangongâ€. It consists of three parts: the perception platform, the machine learning platform and the deep learning platform. Mainly around the openness and output of artificial intelligence technology capabilities. At this point, Baidu Cloud has realized the intelligent platform service layout of artificial intelligence, intelligent big data, intelligent multimedia and intelligent Internet of Things.
Perception platform
The sensing platform mainly includes image technology, voice technology and natural language processing, which can be applied to scenarios such as intelligent customer service, identity verification, content review, etc. Application developers can directly call the API for specific scene applications.
Among them, image technology includes text recognition, face recognition, etc., while speech technology includes speech recognition and speech synthesis, voiceprint recognition, etc. Natural language processing includes Baidu's NLP Cloud capabilities.
When explaining the role of this platform, Liu Wei, general manager of Baidu Cloud Business Unit, reminded us that we need to remember a word - smart "Cong". He said that "Cong" is composed of the eyes and ears, and the function of "eyes and ears" is what Baidu perception platform can do - speech recognition, image recognition and natural language synthesis.
Therefore, in his view, compared with the other two platforms, Baidu's perception platform has a feature: since this platform is based on Baidu's own collection of human-computer interaction data to learn and produce models, its API can be directly externally Make a call. Therefore, the user does not need us to understand the technical details of it can be used.
In other words, the user base that the perceptual platform can reach is also the largest among the three platforms.
Machine learning platform
The machine learning platform is a machine learning service hosted by Baidu Cloud, but users who use the machine learning platform must master a large amount of industry data.
On the machine learning platform, Baidu has opened up the data training process and deeply integrated with the Tianji platform (Intelligent Big Data Platform).
The platform has built in more than 20 common machine learning algorithms, supports the industry standard Spark MLlib, and also corresponds to the massive data and common templates inside Baidu.
But using this platform requires users to provide their own data. Liu Wei explained that if it is a matter of one second to solve, users can solve it through the sensing platform; but if you need to predict based on a large number of recurring data, Baidu's machine learning platform is very suitable for such a scenario.
Liu Wei cited an example. If you have a large amount of data in the field of logistics, you can use the machine learning platform to predict which road from A to B will be the shortest and most convenient. This is a problem that is suitable for machine learning platforms.
Deep learning platform
The deep learning platform supports a variety of neural network structures and optimization algorithms as well as custom network configurations. It optimizes computing, storage, architecture, communications, etc., supports multi-core, multi-GPU, multi-machine environments, and its Paddle internal technology has been mature. And open to global developers.
According to Liu Wei, this platform has a large number of neural network algorithms, which are very efficient and flexible. If users study small-scale data, they can do it on a stand-alone basis; but if the amount of data reaches a certain scale, it can run and build a model based on a distributed system; if the amount of data is very large, a cluster of CPUs is needed for learning.
So what kind of users is the deep learning platform suitable for?
Its target user group is definitely narrower than the user base of the perception platform and the machine learning platform. It is more suitable for data scientists and engineers with very high professional level. Users must have a deep understanding of the data model operation to use this data learning platform.
to sum up
The release of the Tianzhi platform marks that Baidu’s brain capabilities will be officially opened to the public through Baidu Yuntianzhi. In the future, Baidu Cloud will provide intelligent solutions for all walks of life through the four intelligent platforms of Tianchi, Tianxiang, Tiangong and Tianzhi, and fully meet the ABC era (cloud computing, artificial intelligence and big data). Of course, the "Tianzhi" released today is only the primary 1.0 version, which means that Baidu's artificial intelligence cloud platform has a long way to go.
Flum Float,Flum Float 3000Puffs,Fruits Flavors Flum Float,Electronic Cigarettes Flum Float
Nanning Nuoxin Technology Co., Ltd. , https://www.nx-vapes.com