Multi-dimensional analysis of the trend of the "deep learning" market

Multi-dimensional analysis of the trend of the "deep learning" market

[[192373]]

Industry Analysis

In 2016, the global deep learning market size was estimated to have reached $227 million. With more and more applications in the autonomous driving and healthcare industries, deep learning should continue to make outstanding contributions to industry growth. Its technical advantages in overcoming data volume, strong computing power, and data storage capacity have made it a rising star in fields such as speech and images that require high data complexity, providing huge research space and value.

The growing amount of data in various industries is also driving industry development. In addition, the huge demand for human-machine interaction has also provided various solution providers with new ways to develop solutions and functions. However, the data required to train neural networks is a challenge for industry growth.

Major companies are investing heavily in combining deep learning technology with their products. In November 2016, SK Telecom announced that they were working with Intel to develop V2X and video recognition technology based on deep learning. In addition, government support and budget increases in this area will also promote growth in the industry in the next few years. For example, China's National Development and Reform Commission has invested heavily in supporting the development of deep learning research laboratories.

Solution Analysis

At present, the development of deep learning is mainly concentrated in the software level. Through SaaS based on deep learning and machine learning technology, it has brought disruptive changes to the entire industry. These solutions are not only the organization and collection of data, but also can extract a lot of useful information from it for prediction and judgment.

On the other hand, the development of algorithms and hardware still has a long way to go, which also drives the development of chips. Under the growing demand, FPGA and application-specific integrated circuits (ASIC) are also being updated rapidly to meet customer needs.

Hardware Analysis

In 2016, GPUs dominated the hardware space and were much faster than other chips. The increasing demand for enhanced graphics content has led to the need for GPUs in deep learning applications.

On the other hand, the increased use of GPUs by large companies for research and development will also increase demand for GPUs. For example, Google announced that it would add GPUs to its cloud machine learning and computing engines in early 2017 to improve the performance of a large number of computing tasks. GPUs are witnessing tremendous development in training deep learning models with neural networks.

When FPGA first entered the field of deep learning in 2016, it only accounted for a small amount of revenue. However, everyone is generally optimistic that it will have greater development and be able to achieve higher efficiency than GPU. Now FPGA is still in its infancy, but we expect it to become an important player in this field.

Industry Application Analysis

Image recognition has gained huge attention in the industry in 2016, with revenue exceeding 40% of the total share. The most widely used application of this technology is Facebook's face recognition function. It is also widely used in the field of pattern recognition of unstructured data, such as voice, text, images and videos.

In addition, in the next eight years, image recognition applications in the medical and security fields will also rapidly promote the development of the industry. The automotive and financial industries will also continue to transform and integrate with high-tech, using technology to further improve operational capabilities and the ability to transform and implement technology, bringing more value to business and users.

Data mining technology has a 5% market share in 2016. Data segmentation prediction for pattern recognition and effective prediction is the main driving force for the growth of this technology. Using data mining technology to make decisions and inferences is bringing disruptive changes to the field of big data analysis.

Terminal application analysis

Deep learning in aerospace and defense accounted for 20% of the total market revenue in 2016, mainly from applications in remote sensing, object detection and positioning, spectral analysis, identification of network anomalies, and malicious code detection. In addition, with the gradual introduction of wearable computing from cockpits to infantry regiments, the demand for general-purpose GPUs has surged.

Aerospace and defense are using deep learning to address defense challenges through embedded platforms running large amounts of data. Through image processing and data mining techniques, these solutions are able to predict and assess future courses of action. For example, the U.S. Department of Homeland Security uses deep learning in its integrated environmental analysis and simulation project to assess possible future events.

The automotive industry also accounted for a significant share of the deep learning industry revenue last year. This is because the automotive industry is now transforming from private ownership to a shared economy. Automakers are beginning to realize the significance of self-driving cars and are beginning to incorporate deep learning into their ecosystems. Audi uses deep learning algorithms in its camera-related technology to recognize traffic signs by features and shapes.

Regional Analysis

Due to the increase in investment in artificial intelligence and neural networks, the North American market accounted for more than 45% of the total revenue in 2016. This growth trend will continue in the foreseeable future. The North American market is very receptive to cutting-edge technology, which has also led to a rapid adoption of deep learning technology by companies in the region.

On the other hand, increasing government support has also stimulated the development of this field. The US federal government has established a professional committee for artificial intelligence and machine learning, which has also enabled the industry to develop rapidly.

Europe is also paying more and more attention to artificial intelligence, which has further expanded the growth space of the deep learning industry. Especially in the UK, autonomous driving, smart devices and network security are driving the continuous development of the entire industry.

Competitive Analysis

The main players in the market include Nvidia, Intel, Google, and Microsoft. These players are currently expanding their advantages in R&D and market share through continuous mergers and acquisitions. In August 2016, Intel acquired Nervana to obtain its hardware chip platform.

At the same time, many companies are also increasing their investment in the hope of adding deep learning capabilities to their products. In November 2016, GE Healthcare announced a collaboration with the University of California, San Francisco to develop a deep learning algorithm library to improve the efficiency and accuracy of physicians in diagnosing and treating patients.

<<:  Android screenshots and WebView long pictures sharing experience summary

>>:  Build a super fast computer for deep learning for less than $1,000: Continue the exploration of deep learning and cheap hardware!

Recommend

A new direction for battery development? Lithium-sulfur batteries

From mobile phones to computers to cars, lithium ...

Survey shows only 5% of consumers are interested in buying a Fire Phone

At present, the competition in the smartphone mar...

Which WeChat mini program mall development company is best in Guangzhou?

With the continuous upgrading of WeChat mini prog...

What are the main ways to promote apps at home and abroad?

At present, domestic Internet traffic is becoming...

Ask yourself! 15 questions that IT technicians must think about

People in the industry call themselves programmer...

Guess how destructive an egg-sized hailstone is.

Since April, severe convective weather has been v...

Analysis and judgment of the decline in user activity!

User behavior data is a foundation that can help ...