In this series we work towards deep learning. To paint a clear picture of deep learning, however, we first discuss machine learning and neural networks. To explain deep learning clearly, it is important to outline the connections between all these concepts. After the short introduction in the previous article, we now cover these topics in more detail. We start with machine learning. In this introduction we give a definition. What is machine learning?
Machine learning is a broad research field within Artificial Intelligence (AI), which deals with the development of algorithms and techniques with which computers can learn. Machine learning refers to any system where the performance of a machine in performing a task improves by gaining more experience in performing that task. Machine learning, therefore, consists of algorithms that learn thanks to data. It involves using statistical / mathematical techniques to enable computers to learn without being explicitly programmed. Using algorithms that iteratively learn from data, machine learning can find hidden insights without being explicitly programmed where to look.
Text continues below the image.
A commonly used definition of machine learning is: A computer program is said to learn from experience E with respect to task T and performance gauge P, if its performance with tasks in T, as measured by P, improves with experience E. So if you want your program to make predictions, for example, traffic patterns at a busy intersection (task T), you use an algorithm for machine learning with data on historical traffic patterns (experience E) and, if it has learned successfully, then it will perform better in predicting future traffic patterns (performance measure P).
When machine learning models are exposed to new data, they can adapt independently. They learn from previous calculations to produce reliable, repeatable decisions and results. It is important to understand that the ‘learning’ effect is twofold: learning data (known observations) and learning new events (new observations). The latter is really about learning from experience. The goal is to automate these decisions and predictions as much as possible on the basis of self-learning algorithms (i.e. without human intervention).
Machine learning is part of AI and is often used in the development of AI applications, such as Apple’s Siri for speech recognition. Given the attention to the concept, you would almost think that it is something new. However, the first algorithms were used 50 years ago. What makes machine learning so interesting right now is that the world around us has changed: The digital era has led to an explosion of data in all forms and from all regions of the world.
This data, known as Big Data, comes from sources such as social media, internet search engines, e-commerce platforms, online cinemas, etc. This huge amount of data is easily accessible and can be shared via applications such as cloud computing. However, the amount of data that is normally unstructured is so great that it can take decades for people to understand and extract relevant information.
In our blog you can find more English articles about tech.
For those looking to become a Machine Learning Engineer, this hourly calculator might help you know what your rates would be.
You might also like these articles:
Machine learning, neural networks, and deep learning explained
What is AI (Artificial Intelligence): An introduction
What is a neural network: An introduction
What is deep learning: An introduction
