Jing Zhang speed of 350km high-speed rail operators to achieve autopilot
December 31 news, high-speed Beijing-Zhangjiakou Railway (Beijing-Zhangjiakou high-speed rail) December 30 opening and operation of the railway Chongli synchronization put into operation. It is reported that Beijing-Zhangjiakou railway in the world for the first time the number of intelligent revival of 350 kilometers per hour EMU autopilot.
Source: http://www.egsea.com/news/detail?id=541607Samsung will launch a new artificial intelligence platform or "Neon" at CES
According to reports, Samsung will launch or artificial intelligence platform called "Neon" at the upcoming CES. The platform developed by US-based technology and Samsung's Advanced Research Laboratory at (Samsung youngest director) leader Pranav Mistry of the president. The official Twitter account of the department recently released a tweet said: "Neon = artificial human." Artificial human AI is a new concept, referring to accompany designed hardware or software, including robotics and digital pets.
Source: http://www.koreaherald.com/view.php?ud=20191230000574Stanford scientists have won new breakthroughs in the field of brain-computer interface
December 31 news, recently, neurosurgery specialists and engineers recent paper from Stanford University, made an unexpected discovery, the future may allow researchers to help people lose the ability to speak the language recovery in the field of brain-computer interface. They used brain-computer interface to decode certain neurons in brain regions "hand and arm movement" signal, so that a few paralyzed people successfully achieved rapid precise idea of typing. The researchers hope that on the basis of this finding, the future can create an implantable brain medical equipment to help people lose the ability to speak the language ability to recover.
Source: https://tech.sina.com.cn/d/f/2019-12-31/doc-iihnzahk1083923.shtmlHuawei proposed causality discovery algorithm based on reinforcement learning
Huawei Noah's Ark lab out of a paper received ICLR 2020 "CAUSAL DISCOVERY WITH REINFORCEMENT LEARNING " Introduction: Huawei Noah's Ark causal laboratory research team will apply reinforcement learning method to discover causal scoring algorithm, based on attention through self encoder-decoder neural network model to explore the relationship between force mechanism of the data, the causal structure of the binding conditions, and the use of reinforcement learning strategies gradient of the neural network training parameters, finally get a causal graph structure. In the academic world to some common data model, the method performance in the mid-size figure is superior to other methods, including traditional and recent causality discovery algorithm based on gradient algorithm. Meanwhile, the method is very flexible, and can be used in conjunction with any of the scoring function.
Source: https://www.jiqizhixin.com/dailies/c492ece3-7c29-4650-bde0-be112657961fCloud test data: Before AI business, we must first teach them cognitive world
With the application of AI technology in the sinking scenes, AI algorithms corporate landing requirements higher and higher, and vertical fine custom data is particularly important. 2018, China artificial intelligence underlying data services market size of 2.586 billion yuan, of which data resources customized services accounted for 86%. Testin cloud measured its AI data service brand "cloud measured data" appears, it is a typical case. By custom building scene, AI companies or departments to help build barriers core data, provide artificial scene data landing most needed, both to ensure that customers get differential advantage, but also one of the core competitiveness of the cloud measured data.
Source: https://mp.weixin.qq.com/s/ny5D0bRrf2glQQ2RTG5FrA
I wish you all a Happy New Year!
This daily measurement data provided by the cloud finishing, material copyrighted by the original report. Cloud measured data are professional service providers of artificial intelligence data, if you have a data acquisition and labeling requirements, please contact.