Page 42 - DCAP506_ARTIFICIAL_INTELLIGENCE
P. 42

Artificial Intelligence




                    Notes          There are two fundamental types of change:
                                      Relevant Change: examine the changes made by an action
                                      Irrelevant Change: do not examine facts that are not related to the task at hand
                                   Facts may be examined utilizing two levels:
                                      Semantic Level: This level interprets what type of information is being inspected. Solutions
                                       should become understandable  by the  suppositions of  how an  object should  behave.
                                       There are believers in a wholly semantic approach who consider that accurate information
                                       can be reached by means of meaning. However, this hypothesis has yet to be confirmed.
                                      Syntactic Level: This level just decides in which format the information should be examined.
                                       That is, it generates solutions depending on the structure and patterns of facts.
                                   When examining the facts, numerous problems can happen:
                                      At times an implication can be missed.
                                      Considering all facts and all their succeeding side effects is time-consuming.

                                      Some facts are needlessly inspected when they are unneeded.




                                      Task  Make distinction between semantic level and syntactic level.
                                   3.3.1 Problems Related to the Frame Problem


                                   The Qualification Problem
                                   The Qualification Problem was initiated by John McCarthy. It recommends that one is never
                                   totally positive if a particular rule will work. It also proposes that the robot does not essentially
                                   know which rules to disregard in a specified situation. Modifications in the environment can
                                   “confuse” the robot as certain rules will turn out to be obsolete and new rules will be essential
                                   before they exist.

                                   The Representational Problem

                                   The Representational Problem is the complexity of developing truths concerning the current
                                   environment. For instance, how can one program the thoughts of up and  down? These  are
                                   relative to each other, and can not be just described by direction. To incompletely correct this
                                   problem, successor-state axioms are accessed. These axioms demonstrate all the true and false
                                   potentials of a rule.

                                   The Inferential Problem

                                   Difficulty with inspecting the methods by which the world is reviewed is the Inferential Problem.
                                   There are two types of purposes. The General Purpose is to check the complete world of things
                                   that are changeable. The particular Purpose is to only examine actions that can amend over a
                                   little area of environments.

                                   The Ramification Problem

                                   This problem illustrates how an action can cause deviations inside its environment.






          36                                LOVELY PROFESSIONAL UNIVERSITY
   37   38   39   40   41   42   43   44   45   46   47