Artificial Neural Network: The Brain Behind Today’s Smart Technology. Technology will continue to expand its reach into every sector of the economy. Scientific advancement is directly linked to future job growth and employment in the industry, defense, transportation, agriculture and healthcare sectors. Christopher “Kit” Bond, a former U.S. Senator, said that. Senator. Technology is now an integral part of our lives, be it in business, personal, or professional life. Nearly every day, billions of people use social media, search and virtual assistant apps. Ever wondered how Siri and Alexa work? How Facebook recognizes your photos? It’s normal to be confused about how these technologies work. Let’s dive into the details to find out more about how this stuff works. We are now in the age of Artificial Neural Network powered AI. The Artificial Neural Network (ANN) and How It Works Artificial Neural Network Frank Rosenblatt, a psychologist, invented the first Artificial Neural Network (ANN) in 1958. He called it “Perceptron”. It was created to mimic the human brain’s ability to process and analyze visual data such as words and pictures. It allows computers, devices and software to learn by themselves and can recognize patterns and solve problems that are too difficult or impossible for humans. Google DeepMind is an example of an advanced neural network that will drive machine learning in the future. Google DeepMind created an AI system called AlphaGo, which plays Chinese board game Go. AlphaGo beat Lee Sedol (18-time world champion) in Go by learning through an artificial neural net. This was a shocking score of 4-1 last 2016 It caused a lot of buzz in the tech industry. (Photo source: Forbes, “Blue-Collar Revenge”: AI’s Rise Will Create a New Professional Class. ANN can learn from its past experiences or rely on the patterns and data it has processed. AlphaGo beat the grandmaster with its “artificial neurons” using this method. Google DeepMind announced a shift in its AI focus to science and games. “From building AI agents that play games to building AI agents with real world impact, especially in science areas like biology,” said the company. Google DeepMind plans to use its AI products and research in order to harness advances in healthcare, physics, and global warming. Neurons and Synapses https://9to5google.com/guides/deepmind/ Synthetic synapses connect artificial neurons. Synapses are a connector that allows information to be transferred from one neuron into another. ANN’s synapses and neurons can perform calculations and create neuromorphic chip to understand images, sounds, and respond to changes. They all work together to produce one output report. The Three Layers


(Image taken from Scientific American, Unveiling Deep Learning’s Hidden Layers) -The Input Layer’s function is to process all input information and pass it on to the Hidden Layer. A hidden layer is made up of multiple layers that filter data and apply a different transformation on the input data so that it can be used by the Output Layer. -The Output Layer of an ANN, which receives data from the previous layers and transmits the designed data. Summary Analogy: Input Layers are like researchers. They gather, analyze, and interpret all the necessary raw data for a research project. Hidden layers are like a vintner. They extract the best quality grapes from which to make wine. The Output Layer is similar to corporate secretaries. They sometimes receive instructions and messages from various callers, such as clients, employees, and business people, and inform their bosses. Learning how the Neural Networks Learn. ANN can function in the same way as the human brain and can learn the same way that we learn. Howard Rheingold, an American critic and teacher, is well-known for his work on the political, social and cultural implications of modern technology. He stated that the neural network is not an algorithm. It is a type of technology that has weights that can be adjusted to make it learn. It is taught through trials. Source: https://www.goeduhub.com/It’s a fact that the neural network can operate and improve its performance after “teaching” it but it needs to undergo some process of learning to acquire information and be fami