In September 1955, John McCarthy, a young assistant professor of mathematics at Dartmouth College, proposed that "every aspect of learning or any other element of intelligence may, in principle, be so accurately described that a machine can made to simulate it. "
McCarthy called this new field of study" artificial intelligence "and suggested that a two-month effort of a group of 10 researchers could make significant progress in developing machines that could" use language, form abstractions and concepts, solve problems that are now reserved for people and improve. "
At that time, the researchers thought optimally that we would soon think that machines would do any work a human could do. Now, more than six decades later, advances in computer science and robotics have helped us automate many of the tasks that previously required people's physical and cognitive work.
But true artificial intelligence, as McCarthy thought of it, continues to lead us.
What exactly is AI?
A major challenge with artificial intelligence is that it is a broad concept, and there is no clear agreement on its definition.
As mentioned, McCarthy suggested that I solve problems that man does: "The ultimate effort is to make the computer programs that can solve problems and achieve goals in the world and people." McCarthy said .
Andrew Moore computer science dean at Carnegie Mellon University, gave a more modern definition of the term in an interview with 201
But our understanding of "human intelligence" and our expectations of technology are constantly evolving. Zachary Lipton, editor of Approximately Correct describes the term AI as "aspirational, a moving target based on the abilities that humans have but which machines do not". In other words, the things we ask for AI change over time.
For example, in the 1950s, scientists saw chess and controls as major challenges for artificial intelligence. But today, very few would consider chess players to be AI. Computers already handle much more complicated problems, including detecting cancer, driving cars and treating voice commands.
Narrow AI vs. General AI
The first generation of AI researchers and visionaries believed that we could eventually create human intelligence.
But several decades of AI research have shown that replication of the complex problem-solving and abstract thinking on the human brain is extremely difficult. For one thing, we humans are good at generalizing knowledge and applying concepts that we learn in one field to another. We can also make relatively reliable decisions based on intuition and with little information. Over the years, human level AI has become known as artificial general intelligence (AGI) or strong AI.
The initial hype and the excitement around AI attracted interest and funding from authorities and large companies. But it soon became apparent that human level intelligence was not in the vicinity of early perceptions, and the researchers were hard pressed to reproduce the basic functions of the human mind. In the 1970s, unfulfilled promises and expectations eventually led to the "AI winter", a long period during which public interest and funding in the AI was dampened.
It took many years of innovation and a revolution in deep-learning technology to revive interest in AI. But even now, despite huge advances in artificial intelligence, none of the current AI methods can solve problems just as the human mind does, and most experts believe that AGI is at least decades away.
Today's Mathletes – AI Math Equations "border =" 0 "class =" center "src =" https://assets.pcmag.com/media/images/631857-today- s-mathletes-ai-math-equations.jpg? thumb = y & width = 980 & height = 456 "/>
The flip side, narrow or weak AI is not intended to Reproduce the functionality of the human brain, and instead focus on optimizing a single task: Narrow AI has already found many real-world applications, such as recognizing faces, converting audio to text, recommending videos on YouTube, and displaying personal content in the Facebook news feed.
Many scientists believe that we will eventually create AGI, but some have a dystopic view of the age of thinking machines . In 2014, the well-known English physicist Stephen Hawking described AI as an existential threat to humanity, warning "complete artificial intelligence could spell the end of humanity ".
In 2015, Y Combinator President Sam Altman and Tesla CEO Elon Musk, two other believers in AGI, participated in OpenAI, a nonprofit research lab aimed at creating artificial general intelligence in a way that benefits all humanity. (Musk has since resigned.)
Others believe that artificial general intelligence is a pointless goal. "We do not need to duplicate people. Therefore, I focus on having tools to help us instead of duplicating what we already know how to do. We want people and machines to be partners and do something they cannot do on their own hand " says Peter Norvig research director at Google.
Researchers like Norvig believe that narrow AI can help automate repetitive and arduous tasks and help people become more productive. For example, doctors may use AI algorithms to examine high-speed x-ray scans so they can see more patients. Another example of narrow AI is to fight cyberthreats: Security analysts can use AI to find signals of data breaches in gigabytes of data transmitted through corporate networks.
Rule-based AI vs. Machine Learning
Early AI Creation efforts were focused on transforming human knowledge and intelligence into static rules. Programmers must carefully write code (if-then statements) for each rule that defined AI's behavior. The advantage of rule-based AI, which later became known as "good old-fashioned artificial intelligence" (GOFAI), is that people have full control over the design and behavior of the system they develop.
Rule-based AI is still very popular in fields where the rules are clearcut. One example is video games where developers want AI to deliver a predictable user experience.
The problem with GOFAI is that unlike McCarthy's original premise, we cannot exactly describe all aspects of learning and behavior in ways that can be transformed into computer rules. For example, one defines logical rules to recognize voices and images – a complex event that people perform instinctively – is an area where classic AI has historically struggled.
AI Future An alternative approach to creating artificial intelligence is machine learning. Instead of developing rules for AI manually, engineers "engineer" "their" models by giving them a massive amount of samples. The machine learning algorithm analyzes and finds patterns in training data and then develops its own behavior. For example, a machine learning model can train on large volumes of historical sales data for a company and then make sales forecasts.
Deep learning, a subset of machine learning has become very popular in recent years. It is especially good to handle unstructured data, eg. images, video, audio and text documents. For example, you can create a deep-coded image encoder and train it on millions of available tagged images, such as ImageNet dataset . The trained AI model will be able to recognize objects in images with accuracy that often surpasses humans. Advances in deep learning have driven AI to many complicated and critical domains, such as medicine, self-driving cars and education.
One of the challenges of deep learning models is that they develop their own behavior based on training data, making them complex and opaque . Often, deep-learning experts find it difficult to explain the decisions and the inner effects of the AI models they create.
What are examples of artificial intelligence?
Here are some of the ways AI
Self-driving cars: Advances in artificial intelligence have brought us very close to making the decades long dream of autonomous driving reality. AI algorithms are one of the main components that enable self-driving cars to sense the environment, to take food from cameras installed around the vehicle and to detect objects such as roads, traffic signs, other cars and people.
Digital Assistants and Smart Speakers: Siri, Alexa, Cortana and Google Assistant use artificial intelligence to convert spoken words into text and map text to specific commands. AI helps digital assistants feel different shades in spoken language and synthesize human voices.
Translation: For many decades, text between different languages was a pain point for computers. But deep learning has helped create a revolution in services like Google Translate. To be clear, AI still has a long way to go before it manages human language, but so far, progress is spectacular.
Face Recognition: Face Recognition is one of the most popular applications of artificial intelligence. It has many uses, including unlocking your phone, paying with your face and detecting intruders in your home. But the increasing availability of facial recognition technology has also caused concern for privacy, security and civic freedoms.
Medicine: From discovering skin cancer and analyzing X-ray and MRI scans to provide personal health tips and managing the entire health care system, artificial intelligence becomes a key health and medical assessor. AI will not replace your doctor, but it can help improve better health care services, especially in poor areas where AI-powered healthcare professionals can take some of the burden on the shoulders of the few general practitioners who need to serve large populations.
The future of AI
In our quest to crack AI's code and create thinking machines, we have learned a lot about the importance of intelligence and reasoning. And thanks to advances in AI, we perform missions with our computers that were once considered the exclusive domain of the human brain.
Some of the emerging areas where AI gives rise to music and art where AI algorithms manifest their own unique form of creativity. There is also hope that AI will help combat climate change care for the elderly and eventually create a utopian future where people do not have to work at all .
It is also fear that AI will lead to masslessness disrupting the economic balance triggering another world war and eventually driving people into slavery.
We still do not know which direction AI will take. But as the science and technology of artificial intelligence continues to improve steadily, our expectations and definition of AI will change, and what we believe AI today can become the everyday functions of tomorrow's computer.