When Machine Learning was first conceived, it was not clear why this technology was so important. But in 1943, the artificial neural networks were developed, based on a new idea called connectionism. The idea was that neural pathways strengthen over time and this would help us understand how the human brain functions. Today, machine learning has made the world a better place, and there are many examples of how it’s used.
The term “machine learning” was coined by computer scientist Arthur Samuels. At the time, Samuels was working on an algorithm for a checkers game. The board was too large to encode all the moves, but he managed to teach the algorithm to predict several moves ahead. The algorithm proved to be quite competitive with humans and anticipated the breakthroughs in AI, including AlphaGo Zero, which have surpassed all human chess players.
A year later, Gerald Dejong published a research paper on the Neocognitron, an artificial neural network with multiple layers. This research inspired the convolution neural network for deep learning. In the early 1970s, Gerald DeJong introduced explanation-based learning (XBL), which allows ML algorithms to ignore irrelevant data. In 1984, Terrence Sejnowski invented the NETtalk computer program, the first ever artificial neural network. The program learned to speak like a baby using written English texts and phonetic transcriptions.
The development of artificial intelligence and computer gaming has led to the development of machine learning. The first computer learning program was developed by Arthur Samuel and IBM in the 1940s. The IBM computer studied moves that led to winning by analyzing and putting them into a computer program. Later, Frank Rosenblatt created the first neural network for computers, a precursor to today’s neural networks. As AI advances, more uses of artificial intelligence are being developed, including facial recognition and image recognition.
Arthur Samuel coined the term “machine learning” and conducted research on the game of checkers. In 1962, Robert Nealey played checkers on an IBM 7094 computer and lost, which was hailed as a major step forward in artificial intelligence. The development of processing and storage will lead to innovative products. That’s what the next decade will bring. And while it’s unclear when this technology will be able to do, its potential is undeniable.
In 1970, Marvin Minsky predicted the development of machine intelligence. He aimed to make computers capable of reasoning abstractly. But the basic proof of principle was there, but the technology was years away from reaching the end goal. It took a decade before it became an actual reality. As patience dwindled and funding began to dry up, research was stagnant. This slowdown lasted for ten years.
Deep learning is an important part of machine learning, which shifts from a knowledge-based approach to one based on data. Researchers create programs that help computers analyze large amounts of data and make conclusions based on these results. In 2006, IBM’s Deep Blue beat the world champion at chess with the help of machine learning. A few years later, Microsoft releases the DeepFace algorithm, which can detect 20 human facial features and even identify human faces in pictures.
As computers became faster, cheaper, and more powerful, people learned how to apply algorithms to a variety of problems. This process of self-learning algorithms has practical applications in a number of industries. When used correctly, machine learning can uncover optimal solutions for practical problems. And that’s where its real business value lies. So, when Machine Learning Was Invented? – The Evolution of Artificial Intelligence
One example of how machine learning can solve everyday tasks is the detection of cancer from computer tomography imaging. Researchers collect as many CT images as possible, some of which show cancerous cells while others contain healthy tissue. They then train the algorithm to identify the difference between the two. Eventually, it would be able to recognize the differences between these types of images and make the right decision for a patient. That’s where it gets interesting!