Presentación realizada para la materia de Inteligencia Artificial de la Licenciatura en Ingeniería en Mecatrónica de la Benemérita Universidad Autónoma de Puebla en el periodo de Primavera 2017, que contiene la definición de inteligencia artificial, sus disciplinas y sus diferentes enfoques
2. DEFINITION
Thinking Humanly
• Effort to make machines with
minds.
• Automation of activities as
decisión making, problema
solving, learning.
Thinking Rationally
• Study of mental faculties with
computational models.
• Study of computation to make it
posible to perceive, reason and
act.
Acting Humanly
• Creation of machines that
performe functinons that requiere
intelligence.
• How to make computers do things
at which people are better.
Acting Rationally
• Study of the design of intelligent
agents.
• Concerned with intelligent
behavior in artefacts.
3. ACTING HUMANLY
THETURINGTEST APPROACH
TheTuringTest: to provide a satisfactory operational definition of intelligence
A human interrogator, after posing some write questions, cannot tell wheter the
responses come from a person or a computer.
Physical stimulation of a person is unnecessary for intelligence
TotalTuring test: includes a video signal
Test the subject’s perceptual abilities
5. THINKING HUMANLY
THE COGNITIVE MODELING APPROACH
Determine how humans think through
Introspection
Psychological experiments
Brain imaging
Cognitive science
Brings together computer models from AI and experimental techniques from
psychology to constructe precise and testable theories of the human mind
“an algorithm performs well on a task and that it is therefore a good model of
human performance, or viceversa the cognitive modeling approach”
6. THINKING RATIONALLY
THE “LAWS OFTHOUGHT” APPROACH
Syllogism: patterns of argument structures that always yielded correct
conclusions when given correct premises
Govern the operation of mind
Logic: study of syllogisms
Logicist tradition within AI hopes to build programs… to create intelligent systems
Solve any solvable problem described in logical notation (if no solution exists, the
program might loop forever)
Obstacles
It’s not easy to take informal knowledge and state it in the formal terms required by logical
notation
There is a big difference between solving a problema “in principle” and solving it in practice
7. ACTING RATIONALLY
THE RATIONAL AGENT APPROACH
Agent: something that acts
Computer agent
Operate autonomously
Perceive its environment
Adapt to change
Create and persue goals
Rational agent
Acts to achieve the best outcome or the best expected outcome (when there is uncertainty)
Making correct inferences is sometimes part of being a rational agent, but it’s not all of rationallity, sometimes there is no
correct thing to do, but something must still be donde.
Advantages
More general than laws of thought
More amenable to scientific approaches
The standard of rationality is mathematically well defined and completely general and can be “unpacked”
Limited rationallity: acting appropriately when there is not enough time to do all the computations one might like