News classification
Contact us
- Add: No. 9, North Fourth Ring Road, Haidian District, Beijing. It mainly includes face recognition, living detection, ID card recognition, bank card recognition, business card recognition, license plate recognition, OCR recognition, and intelligent recognition technology.
- Tel: 13146317170 廖经理
- Fax:
- Email: 398017534@qq.com
A11 neural network engine helps apple AI leap
A11 neural network engine helps apple AI leap
With a new generation of apple iPhone smart mobile phone iPhone 8, iPhone 8 Plus and iPhone X release, any information we want to know all the original now almost all have been crystal clear, including the number of retail price, the new name, type, there are many new functions and so on, the most of the information in the conference before it has been broke out the. Still, the new features of iPhone, X, facial recognition technology, and AR augmented reality are still very impressive.
In fact, A11 performance has long been a leader can be expected, but the "most intelligent" and where? In fact, this has a lot to do with the A11 integrated neural network engine. Although this part of the apple press conference with a pen, but low-key publicity, but it is undeniable that the A11 bionic become SoC mobile chip Apple first neural network engine, the neural network engine is specially used for processing unit operation AI of artificial intelligence, this apple A11 named by "bionic" especially the technical characteristics of AI.
Privacy protection in conflict with the development of AI artificial intelligence
Before discussing the neural network engine, let's begin with the apple AI AI process. As we all know, apple is an extremely privacy conscious company, and many examples have been proven before. Cook has even declared that apple is more secure than CIA in the United states.
Therefore, a large number of analysts believe that it is the privacy of the protection of the plot, so that Apple lost the initiative to AI artificial intelligence, and ultimately lead to today's AI artificial intelligence competition behind. Apple not only lags behind other tech giants in making its devices more user-friendly, but also has a low profile in artificial intelligence and cloud computing, and has been marginalized for a long time. Why do you say so?
Deep learning is a branch of artificial intelligence AI, usually need to collect large amounts of data, and these data aggregation in the cloud, in order to understand the depth, gain insight, but these practices with apple strict stand on data protection is not consistent. Apple in iOS system FaceTime and iMessage applications are using end-to-end encryption technology, and do not have any personal data or information about the user's maintenance or management, which is also consistent with the firm views on consumer Apple data protection.
While Apple's position is commendable, it has made its AI development difficult, as apple itself does not collect the big data needed to boost its AI artificial intelligence, until now.
AI's new beginning is not the beginning
In today's mobile Internet era, data explosion, the emergence of big data to make AI artificial intelligence become better and better. Although many mobile device data can be further analyzed in cloud computing, there are powerful AI computing solutions such as Google, TPU, and NVIDIA Volta. But the problem is that the data you need to spend some time to get to the cloud, and Apple will not consider the privacy of data to the cloud, so let the computing performance of a mobile device to provide a bit closer to the cloud hardware is necessary.
Apple explained at the conference, there is a AI artificial intelligence called machine learning, that is, the device through observation way of learning. Therefore, a mobile SoC chip, there should be a AI processing unit specializing in virtual neurons and deep learning, and the neural network engine is designed for machine learning and the development of hardware, it not only can perform high speed operation for the neural network, but also has outstanding efficiency.
In short, through the neural network engine can undertake the tasks of CPU and GPU, greatly enhance the efficiency of chip computing, less energy consumption, faster completion of more tasks. The A11 bionic neural network engine, Apple's own two core dual core design, designed for specific artificial intelligence AI machine learning algorithms, real-time processing, operation times up to 600 billion times per second.
Apple said that the neural network can run in the CPU main processing unit or GPU graphics processing unit, but for this type of neural network programming model, using custom chips for these applications, to perform the same task, it will be more efficient than using graphics engine. The mystery of the neural network engine is that it can handle matrix multiplication and floating-point processing, and share specific tasks outside of CPU and GPU, enabling a substantial improvement in hardware performance.
In fact, Apple has long been well aware of the trend of mobile AI processing units. A few days ago, said Sjonny Sloki, senior vice president of Apple hardware technology interview, 3 years ago published equipped with A8 chip iPhone 6, Apple has been in the development of A11 bionic chip, however, about AI and machine learning topic three years ago, the mobile industry is not much. Sjonny Sloki highlighted, "the built-in neural network engine is apple in 3."
However, there is a possibility that Apple has not featured in the press conference or publicity, and many people may ignore its existence, that is, the powerful AI artificial intelligence characteristics.
In fact, A11 performance has long been a leader can be expected, but the "most intelligent" and where? In fact, this has a lot to do with the A11 integrated neural network engine. Although this part of the apple press conference with a pen, but low-key publicity, but it is undeniable that the A11 bionic become SoC mobile chip Apple first neural network engine, the neural network engine is specially used for processing unit operation AI of artificial intelligence, this apple A11 named by "bionic" especially the technical characteristics of AI.
Privacy protection in conflict with the development of AI artificial intelligence
Before discussing the neural network engine, let's begin with the apple AI AI process. As we all know, apple is an extremely privacy conscious company, and many examples have been proven before. Cook has even declared that apple is more secure than CIA in the United states.
Therefore, a large number of analysts believe that it is the privacy of the protection of the plot, so that Apple lost the initiative to AI artificial intelligence, and ultimately lead to today's AI artificial intelligence competition behind. Apple not only lags behind other tech giants in making its devices more user-friendly, but also has a low profile in artificial intelligence and cloud computing, and has been marginalized for a long time. Why do you say so?
Deep learning is a branch of artificial intelligence AI, usually need to collect large amounts of data, and these data aggregation in the cloud, in order to understand the depth, gain insight, but these practices with apple strict stand on data protection is not consistent. Apple in iOS system FaceTime and iMessage applications are using end-to-end encryption technology, and do not have any personal data or information about the user's maintenance or management, which is also consistent with the firm views on consumer Apple data protection.
While Apple's position is commendable, it has made its AI development difficult, as apple itself does not collect the big data needed to boost its AI artificial intelligence, until now.
AI's new beginning is not the beginning
In today's mobile Internet era, data explosion, the emergence of big data to make AI artificial intelligence become better and better. Although many mobile device data can be further analyzed in cloud computing, there are powerful AI computing solutions such as Google, TPU, and NVIDIA Volta. But the problem is that the data you need to spend some time to get to the cloud, and Apple will not consider the privacy of data to the cloud, so let the computing performance of a mobile device to provide a bit closer to the cloud hardware is necessary.
Apple explained at the conference, there is a AI artificial intelligence called machine learning, that is, the device through observation way of learning. Therefore, a mobile SoC chip, there should be a AI processing unit specializing in virtual neurons and deep learning, and the neural network engine is designed for machine learning and the development of hardware, it not only can perform high speed operation for the neural network, but also has outstanding efficiency.
In short, through the neural network engine can undertake the tasks of CPU and GPU, greatly enhance the efficiency of chip computing, less energy consumption, faster completion of more tasks. The A11 bionic neural network engine, Apple's own two core dual core design, designed for specific artificial intelligence AI machine learning algorithms, real-time processing, operation times up to 600 billion times per second.
Apple said that the neural network can run in the CPU main processing unit or GPU graphics processing unit, but for this type of neural network programming model, using custom chips for these applications, to perform the same task, it will be more efficient than using graphics engine. The mystery of the neural network engine is that it can handle matrix multiplication and floating-point processing, and share specific tasks outside of CPU and GPU, enabling a substantial improvement in hardware performance.
In fact, Apple has long been well aware of the trend of mobile AI processing units. A few days ago, said Sjonny Sloki, senior vice president of Apple hardware technology interview, 3 years ago published equipped with A8 chip iPhone 6, Apple has been in the development of A11 bionic chip, however, about AI and machine learning topic three years ago, the mobile industry is not much. Sjonny Sloki highlighted, "the built-in neural network engine is apple in 3."