Ask INDIAai: Why does deep learning have the name “deep”? – INDIAai

In this series, you can ask questions and get your doubts cleared. The expert team from INDIAai will answer your questions.Send your questions to: [email protected]¬†How does deep learning work? Why does it have the name “deep”? – Sinduja Sunder, Chennai.Deep learning is a subset of machine learning. A neural network with three or more layers is essentially what deep learning is. These neural networks mimic how the human brain works, even though they aren’t very good at it. It lets them “learn” from large amounts of data. A neural network with only one layer can still make rough predictions, but adding more hidden layers can help to optimize and refine for accuracy.Deep learning is at the heart of many applications and services that use artificial intelligence (AI) to improve automation by doing analytical and physical tasks without human help. Deep learning is used in everyday products and services like digital assistants, voice-enabled TV remotes, and credit card fraud detection. It is also used in new technologies (such as self-driving cars).Deep Learning gets its name from the fact that we add more “Layers” to learn from the data. If you don’t already know, when a deep learning model learns, it just changes the weights using an optimization function. A Layer is a row of so-called “Neurons” in the middle.Why do we use Machine Learning? Why is it essential? – Parvathi S, Palakad.With machine learning, a user can give a computer algorithm a vast amount of data and have the computer analyze the data and make decisions and suggestions based on the data.All businesses need data to run. Making decisions based on data is becoming increasingly important to keep up with the competition or fall further behind. Machine learning can be the key to unlocking the value of corporate and customer data and making decisions that keep a company ahead of the competition.Can AI predict natural disasters? – Gopinath, DelhiYes, scientists have figured out how big a tsunami will be even when they don’t know much about the earthquake that caused it. Together, we can make predictions using Global Navigation Satellite System (GNSS) data and artificial intelligence (AI).What is responsible AI? Can you give an example? – Sai Aadvik, PuneResponsible AI is designing, developing, and deploying AI to empower employees and businesses while positively impacting customers and society, allowing companies to build trust and scale AI confidently.A self-driving car, for example, can collect images from sensors. A machine learning model can use these images to make predictions (e.g. the object in front of us is a tree). Then, the car uses these predictions to make decisions (e.g. turn left to avoid the tree).Their principles emphasize fairness, transparency and explainability, humanity, privacy and security.What is computer vision? Is computer vision and CNN the same thing? – Monisha Kesavan, HyderabadComputer vision is an interdisciplinary field of science that looks at how computers can understand complex things from digital pictures or videos. From an engineering point of view, it tries to figure out how the human visual system works and find ways to automate tasks that it can do.CNN is a deep learning network for computer vision that can recognize and classify different picture parts. CNN’s design was influenced by how the visual cortex is set up and what it does. It is made to look like how neurons in the brain connect.Facial recognition applications, which let computers match pictures of people’s faces to their names, also use computer vision in a big way. Algorithms for computer vision can find facial features in images and compare them to facial profiles in a database.How does natural language processing work? How is it used in real life? – Ajith Kumar, Vellore.In natural language processing, human language is broken up into pieces so that the grammatical structure of sentences and the meaning of words can be analyzed and understood in context. It lets computers read and understand written or spoken text just like humans do.Here are a few examples that stand out.Email filters, smart assistants, search results, predictive text, language translation, digital phone calls, data analysis, and text analytics.