Page 174 - DCAP506_ARTIFICIAL_INTELLIGENCE

Artificial Intelligence

Notes Self Assessment

Fill in the blank:
18. Learning by recoding cases method is chiefly used in ......................... learning tasks.

12.12 Learning by Managing Multiple Models

Learning can be performed by handling a set of common models and a set of particular models
kept in a version space. A version space is nothing but a demonstration that is used to get
pertinent information from a set of learning examples.

A version space portrayal includes two trees which harmonize to each other: one symbolize
general model and the other symbolize specific model. Both positive and negative examples are
used to get the two set of models converge on one just-right model. That is, positive examples
simplify specific models and negative examples specialize common models. Eventually, a correct
model that matches only the observed positive examples is attained. Query By Committee
(QBC) is an algorithm which executes this technique so as to obtain knowledge.

Self Assessment

Fill in the blanks:
19. That is, ......................... examples simplify specific models and negative examples specialize
common models.
20. A version space portrayal includes two......................... which harmonize to each other: one
symbolize general model and the other symbolize specific model.

12.13 Summary

 Learning is distinguished into a number of different forms. The simplest is learning by
trial-and-error.
 Rote learning is the fundamental learning activity which is also known as memorization
since the knowledge, without any alteration is, just copied into the knowledge base.

 Learning by taking advice is a straightforward form of learning where knowledge must
be transformed into an operational form before stored in the knowledge base.

 Classification is the procedure of allocating, to a specific input, the name of a class to
which it associates.
 The thought of generating a classification program that can develop its own class definitions
is known as concept learning or induction.

 Winston’s Learning Program is an early structural concept learning program which
functions in a simple blocks world domain.

 The objective of version spaces is to generate a description that is reliable with all positive
examples but no negative examples in the training set.
 In case of decision tree, to categorize a specific input, we begin at the top of the tree and
reply questions until we reach a leaf, where the specification is stored.
 An Explanation-based Learning (EBL) system accepts an example (i.e. a training example)
and illustrates what it learns from the example.

168 LOVELY PROFESSIONAL UNIVERSITY