Page 16 - DCAP506_ARTIFICIAL_INTELLIGENCE
P. 16

Artificial Intelligence




                    Notes          Intelligent Behaviour

                                   This leads us to the question of what represents intelligent behaviour. Some of these tasks and
                                   applications are:
                                      Perception including image recognition and computer vision

                                      Reasoning
                                      Learning
                                      Understanding language including natural language processing, speech processing

                                      Solving problems
                                      Robotics.
                                   1.3.2 Task Domains


                                   Task Domains include:
                                      Formal tasks (matematika, games)
                                      Mundane task (perception, robotics, natural language, common sense, reasoning)

                                      Expert tasks (financial analysis, medical diagnostics, engineering, scientific analysis, dll)

                                   Self Assessment

                                   Fill in the blanks:
                                   7.  When learning the usual range of tasks that we might look forward to an “intelligent
                                       entity” to perform, we require to take into account both “commonplace” tasks in addition
                                       to ......................... tasks.
                                   8.  ......................... tasks  are performed by matter  of routinely by people  and some  other
                                       animals.

                                   1.4 AI Technique

                                   As there are many diverse artificial intelligence techniques that have been generated, with new
                                   ways being produced, a few forms of Artificial Intelligence (AI) have turn out to be more and
                                   more well-liked. Some of the most general techniques comprise the use of neural networks and
                                   the expansion of expert systems. These dissimilar artificial intelligence techniques can be accessed
                                   to build up dissimilar forms of AI, typically depending on the amount of “thinking” the program
                                   can really do, and these are called as either “strong AI” or “weak AI.”
                                   Artificial intelligence techniques are the ways that can be used to generate and produce computer
                                   programs generally observed as forms of artificial intelligence. Usually, artificial intelligence
                                   points to a program that is able to imitate or re-create the consideration processes illustrated by
                                   the human brain. This typically comprises solving problems, making observations or obtaining
                                   input for utilization in analysis or problem solving, and the aptitude to classify and identify
                                   dissimilar objects and the properties of those objects.

                                   There are many dissimilar artificial intelligence methods that can be used by an AI programmer,
                                   though two of the most general are neural networks and expert systems. Neural networks are
                                   computer programs intended about the cognitive processes utilized by the human brain. Basically,
                                   a neural network includes layers of categorization and ways by which objects can be recognized




          10                                LOVELY PROFESSIONAL UNIVERSITY
   11   12   13   14   15   16   17   18   19   20   21