Concept Learning and the General to Specific Ordering

Much of learning involves acquiring general concepts from specific training examples. People, for example, continually learn general concepts or categories such as "bird," "car," "situations in which I should study more in order to pass the exam," etc. Each such concept can be viewed as describing some subset of objects or events defined over a larger set (e.g., the subset of animals that constitute birds). Alternatively, each concept can be thought of as a boolean-valued function defined over this larger set (e.g., a function defined over all animals, whose value is true for birds and false for other animals).

Concept Learning: Inferring a boolean-valued function from training examples of its input and output.

What does a Concept mean?

A concept means the spirit of something, the category of something. For example, as human beings we have learned the concept of Ship. We know what it is, and no matter how many new, high-tech and fancy ships we are shown, we can still classify them as ships, even though we might not have seen those specific models before! We can do this because we have learned the general concept of ship that encompasses infinite number of different kinds of ships that were, are, or will ever be built!

What other concepts can you think of?

• Bird

• Stressful situation (That is right! A concept doesn’t have to be a physical object)

• Cold/hot/humid

On a more technical level we can think of a concept as a Boolean function. As w e knows, every function has a domain (i.e., the inputs) and a range (i.e., the outputs)! So can you think of the domain and range of a Boolean function that represents a concept such as the concept of bird?