Deep learning without mysteries: discover how it works

The relevance of big data continues to grow more and more in our lives. The course will bring you closer to various concepts and techniques in the field of big data studies and machine learning, also known in English as machine learning. One of the big data processing techniques taught in the course is deep learning.

Deep learning is an artificial intelligence tool that is relevant to the development of very powerful technological solutions. In this article, we are going to explore this technique to begin to remove the mystery and show some of the topics that you can continue to learn through the course “Big Data without Mysteries.”

From machine learning to deep learning

Do you remember how you learned to communicate? You were hearing or seeing the same words many times and you formed patterns, to learn to identify words and connect concepts until you knew how to generate a response.

Computers learn similarly. Chatbots like Cortana and virtual assistants like Alexa can “understand” our questions and generate relevant answers thanks to the processing of high volumes of data, be it text data or voice recordings. The virtual assistant is trained using algorithms that expose him to millions and millions of examples in the pronunciation of each word. When we talk about machine learning or machine learning, it generally refers to this process of “training” with data.

The role of neural networks

We as humans have a brain that facilitates this learning process, without our realizing it. Instead, machines rely on certain algorithms to guide them through learning processes, the rules they use to review and make sense of all the data that is taught. We can say that the most important gear of deep learning in neural networks, an automated implementation of statistical analysis. The “mystery” behind neural networks is pure statistics and classical probability theory: non-linear regressions, and Bayesian classification, to name a few, science generated since the 17th century. What neural networks do is iteratively process large volumes of data by changing the weight of these formulas, with enormous automatic capacity.

Applications that improve lives

Given its complexity and the “heavy” (in terms of bytes) of processing media such as images and sounds, it had not been possible to advance in the use of algorithms to process this material until relatively recently. However, deep learning has managed to radically change this reality by being able to process this data at very high volumes and speed, thus evolving the recognition of images, audio, and video for artificial intelligence tasks. Deep learning algorithms manage to segment both images and pieces of audio into their smallest components (such as pixels). These are processed iteratively over several rounds, each one enhancing the machine’s ability to recognize matter similar to what it “saw” or “heard” previously.

As an example, the María de Dios Oncology Hospital in Porto Alegre [1] is using Watson-IBM technology to process large volumes of text and images inpatient medical records to suggest personalized treatments according to each particular case.

Deep learning is just one tool among many that deal with big data. Sign up for the ” Big Data without Mysteries ” course now to reveal this and other technologies that are revolutionizing the world!

Leave a Comment