Computer programs can show different forms of artificial intelligence. This includes machine learning the most effective ones, and the ability to use that knowledge to respond in new ways in the future. Human intelligence reflects the learning ability of our brain. Computer systems that behave like humans use artificial intelligence. This means that these systems are controlled by adaptive computer programs. Like humans, computers can learn how to use data and make decisions and decisions based on what they learn. This is called machine learning and is part of a broader field of artificial intelligence.
Arthur Samuel, an engineer at the University of Illinois in the 1940s, decided to program his computer differently. This computer scientist teaches computers to learn for themselves. Instead of programming all possible movements, he received computer advice from a draftsman. He also taught the computer to play only the control. During each game, computers have seen his train and strategy most. Then he used these trains and strategies to play a better game next time. The computer included data bits of information. Samuel completed his first computer program to play this game in a few years. At that time he worked in the IBM laboratory in the picture of New York. In the image processing system, the computer was trained to recognize cats. For example, if a cat is sunbathing by the window. This turned out to be a difficult task. Unlike children, computers were initially difficult to recognize when the cat was in a strange position or when only a part of the cat's body was visible. In 2015 TED talked, Li described about how his team did it. They needed the help of other scientists. About 49,000 volunteers from 127 countries have classified about 1 billion images in about 3 years. In the end, Li's team collected over 62,000 pictures of cats. Some cats were sitting. Others were standing. As computer programs that pass the data of this image, these programs have learned to identify cats in new pictures ready for display. These are formulas or instructions that follow a step-by-step process. For example, algorithmic steps can strain a computer and group it into similar templates. If you like pictures of cats, people will help you analyze the wrong information. In other cases, algorithms can help you find and learn mistakes. One of the more powerful methods of the machine is called "deep training". Organize computational tasks in a system called a neural network (or neural network). The network runs on connected nodes where data is moved and processed. In that sense, these networks are somewhat similar to the human brain. The concept of neural networks was developed by Warren McCullough and Walter Pitts in the 1940s. They later developed this system while working at the Massachusetts Institute of Technology in Cambridge. Neural networks have been out of date for some time. However, they made a big comeback in the 1980s. These serve as the foundation for machine learning systems that are becoming more complex today. In modern deep learning systems, data typically flows only in one direction through nodes (connections). Each level of the system can receive data from child nodes, process that data, and send it to the parent node. As the computer learns, the layers become more complex (deeper). Instead of making simple decisions like in a checker game, deep learning systems see and learn large amounts of data and make decisions based on it. All of these steps are performed within the computer without input from a new person. Artificial intelligence is a tool to help people. Machine learning is now emerging in tools, software, and products designed to make life easier. One example is the programs used in today's smart speakers and streaming services. They will find the music and video trends of your choice and suggest similar trends that you may like. Machine learning is also used to help people solve bigger problems in everything from engineering to medicine. Some machine learning systems use video games as a learning tool. For example, systems developed by Argonne National Laboratory Engineers outside Chicago can test hundreds of various engine designs in the Illinois Islands.
You must be logged in to post a comment.