Computational intelligence, also known as soft computing, is a form of computing modeled on the methods by which humans learn. As computers learn from processes based on logic and science, they become more intelligent. This differs from artificial intelligence in its perspective on imperfection: computational intelligence focuses on the growth of a system and does not use Boolean values (0s and 1s) to achieve learning, where AI does. Computational intelligence uses different branches of science, such as math and logic, to develop machine learning algorithms.
Computational intelligence consists of three main aspects:
Fuzzy logic: unlike artificial intelligence, computational intelligence (CI) enables systems to grow and learn with imperfect, incomplete knowledge. CI may have different degrees of information and process without full understanding. This logic is typically developed by showing images to a computer and allowing it to differentiate between content.
Artificial neural networks: these processes are designed to imitate the brain. The system learns by being given different image examples and being told different things about those images. Having many layers of learning, some more complex than others, is also known as deep learning.
Evolutionary computation: concepts from evolutionary theory, including natural selection and swarm intelligence, can be applied to computing systems. Researchers study animal groups to learn how a group behaves together and on its own, with no clear leadership – yet the group still maintains intelligence. Computing systems, then, could perhaps display intelligence and decision-making without having a human source of leadership.