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
Artificial intelligence
Artificial intelligence
Artificial Intelligence (AI) refers to the ability of a computer to have intelligent ability like a person. It is a frontier integrated subject that combines computer science, statistics, brain neurology and social science. It can replace human recognition, cognition, analysis and decision-making. If you say a word, the machine can recognize words, understand the meaning of your words, stop analysis and dialogue.
Besides, what are the key milestones in understanding the history of AI?
AI formally put forward in 50s and 60s, in 90s, chess champion Kasparov and "deep blue" computer battle, "deep blue" win, this is an important milestone in the development of artificial intelligence. In 2016, Google's AlphaGo won the Korean chess player Li Shishi, which triggered the AI boom again. This year, Tencent launched the go software "amazing art", which represents the technological level of AI in a specific period.
AI burst out from time to time. It is inseparable from the progress of the base equipment and the update of science and technology. From the rise of the personal computer in 70s to the development of the hardware equipment such as the GPU and the heterogeneous computing in 2010, it has laid the foundation for the revival of artificial intelligence.
At the same time, the development of Internet and mobile Internet has brought a series of data to enable AI to progress. Moreover, the computing power has also been greatly innovating from AI, which is traditionally dominated by CPU and GPU. The update of algorithm technology is in the rise of artificial intelligence. The earliest algorithms are conventional statistical algorithms, such as the neural network in 80s, the shallow layer in 90s, the SBM, Boosting, convex, and so on, and so on in the 2000. As the amount of data increases, computing becomes stronger, and the effect of deep learning is bigger and bigger. Since 2011, the rise of deep learning has led to the current high tide of artificial intelligence.
Secondly, what are the research areas and branches of AI?
There are five layers in the field of artificial intelligence research. The bottom is the establishment of base equipment, which includes data and calculation to two parts. The greater the data, the stronger the ability of artificial intelligence. The upper level algorithm is algorithm, such as convolution neural network, LSTM sequence learning, Q-Learning, deep learning and so on. All algorithms are machine learning algorithms. The third level is an important technical direction and problem, such as computer vision, speech engineering, natural language processing, etc. There are also some other similar decision-making systems, such as reinforcement learning, or statistical systems like some large data parsing, which can be generated in machine learning algorithms. The fourth level is detailed technology, such as image recognition, speech recognition, Machine Translation and so on. The top of the industry's processing plan, such as the application of artificial intelligence to financial, medical, Internet, transportation, and games, is the value that we care about.
It is worth mentioning that there is a difference between machine learning and depth learning. Machine learning means that computer algorithms can find information from data like people, and learn some rules. Although deep learning is a kind of machine learning, deep learning is a neural network with depth, which makes the model more complex, so that the knowledge of the model is deepened.
There are three kinds of machine learning. The first kind is no monitoring learning. It refers to the automatic searching law from the information moving, and it is also called "clustering problem". The second is surveillance learning. Surveillance learning refers to a label for history, using model prediction results. If there is a fruit, we will judge whether it is banana or apple according to the shape and color of the fruit, which is an example of surveillance learning. The last category is reinforcement learning, which is a learning method that can be used to support people to make decisions and plans. It is a feedback mechanism that rewards people's actions and behaviors, and promotes learning through this feedback mechanism, which is similar to human learning. So strong chemistry is one of the important directions of the present study.
Besides, what are the key milestones in understanding the history of AI?
AI formally put forward in 50s and 60s, in 90s, chess champion Kasparov and "deep blue" computer battle, "deep blue" win, this is an important milestone in the development of artificial intelligence. In 2016, Google's AlphaGo won the Korean chess player Li Shishi, which triggered the AI boom again. This year, Tencent launched the go software "amazing art", which represents the technological level of AI in a specific period.
AI burst out from time to time. It is inseparable from the progress of the base equipment and the update of science and technology. From the rise of the personal computer in 70s to the development of the hardware equipment such as the GPU and the heterogeneous computing in 2010, it has laid the foundation for the revival of artificial intelligence.
At the same time, the development of Internet and mobile Internet has brought a series of data to enable AI to progress. Moreover, the computing power has also been greatly innovating from AI, which is traditionally dominated by CPU and GPU. The update of algorithm technology is in the rise of artificial intelligence. The earliest algorithms are conventional statistical algorithms, such as the neural network in 80s, the shallow layer in 90s, the SBM, Boosting, convex, and so on, and so on in the 2000. As the amount of data increases, computing becomes stronger, and the effect of deep learning is bigger and bigger. Since 2011, the rise of deep learning has led to the current high tide of artificial intelligence.
Secondly, what are the research areas and branches of AI?
There are five layers in the field of artificial intelligence research. The bottom is the establishment of base equipment, which includes data and calculation to two parts. The greater the data, the stronger the ability of artificial intelligence. The upper level algorithm is algorithm, such as convolution neural network, LSTM sequence learning, Q-Learning, deep learning and so on. All algorithms are machine learning algorithms. The third level is an important technical direction and problem, such as computer vision, speech engineering, natural language processing, etc. There are also some other similar decision-making systems, such as reinforcement learning, or statistical systems like some large data parsing, which can be generated in machine learning algorithms. The fourth level is detailed technology, such as image recognition, speech recognition, Machine Translation and so on. The top of the industry's processing plan, such as the application of artificial intelligence to financial, medical, Internet, transportation, and games, is the value that we care about.
It is worth mentioning that there is a difference between machine learning and depth learning. Machine learning means that computer algorithms can find information from data like people, and learn some rules. Although deep learning is a kind of machine learning, deep learning is a neural network with depth, which makes the model more complex, so that the knowledge of the model is deepened.
There are three kinds of machine learning. The first kind is no monitoring learning. It refers to the automatic searching law from the information moving, and it is also called "clustering problem". The second is surveillance learning. Surveillance learning refers to a label for history, using model prediction results. If there is a fruit, we will judge whether it is banana or apple according to the shape and color of the fruit, which is an example of surveillance learning. The last category is reinforcement learning, which is a learning method that can be used to support people to make decisions and plans. It is a feedback mechanism that rewards people's actions and behaviors, and promotes learning through this feedback mechanism, which is similar to human learning. So strong chemistry is one of the important directions of the present study.