Soft computing is a way of computing that aims to resolve non-linear issues with approximate solutions to help solve the problem. Functional approximation and approximate reasoning abilities form the basis of Soft Computing. This can be achieved by making use of probabilistic models, Fuzzy Logic Systems, Neural Networks and even Evolutionary algorithms. In general, Soft Computing comprises of technologies that are inspired by the ability of human reasoning to handle the level of certainty, precision, and partial truth in any situation. The one advantage of soft computing is that it has the ability handle raw and unstructured data in an effective way. AI and Soft Computing systems have more differences than commonalities. However, both of them have the same goal of effectively using machines to close the gap between human intelligence and machine intelligence.
If you are looking for an online course to learn Artificial Intelligence, check out this Artificial Intelligence Online Course by Intellipaat.