Professional Documents
Culture Documents
AI Is The Part Of Computer Science Concerned With The Design Of Computer Systems Which Exhibit The Attributes Of Human Intelligence Such as: understanding Of Languages, Learning New Information, Reasoning, Solving Problems, Planning, Perception , And Creativity.
The study of how to make computers do things at which, at the moment, people are better(Rich and Knight, 1991) The branch of computer science that is concerned with the automation of intelligent behaviour (Luger and Stubblefield,1993) The study of mental faculties through the use of computational models (Charniak and McDermott, 1985) The exciting new effort to make computers think machines with minds, in the full and literal sense (Haugeland, 1985)
` ` ` ` ` ` `
Use Of Intuition Common Sense - Application Of Knowledge To New Situations Creativity - Creating Something New And Useful Learning From Experience Memory Retention And Recall Goal - Directed Search Human Information Processing Has And Hope To Find In Machines
AI Definitions
Thought processes and reasoning
Behavior
4
Computer Systems Will Be Processing Knowledge As A Result The Application Areas Of Computer Areas Will Increase. New Tools Can Be Built Which Have Commercial Value And Can Improve The Life Style. AI technology can help make computers easier to use with user-friendly interfaces including natural language, and gestures.
The interfaces will allow users to state what they want to accomplish, and automatically determine the actions required satisfying their needs. They will have control over the actions as and when to perform them. Over time, these intelligent systems will build a model of the user's needs and will adjust automatically to an individual's skills and pattern of usage. The field of planning develops algorithms that automatically construct and execute sequences of primitive commands in order to achieve high-level goals.
sequential (or logical) approach that considers only a small portion of the available data at a time parallel (or gestalt) processing looks at data on a global basis
Many tasks which we might reasonably think require intelligence are performed by computers without even thinking Complex Arithmetic
Other tasks that people do without thinking are extremely difficult to automate
Recognizing a Face
8
AI Problems
Mundane tasks correspond to the following AI problems areas: Planning :
The ability to decide on a good sequence of actions to achieve our goals The ability to make sense of what we see
Vision : Robotics:
The ability to move and act in the world, possibly responding to new perceptions Natural Language: The ability to communicate with others in any human language
Mundane tasks are generally much harder to automate
9
Problem oriented ( different problems have to be dealt in a different way) Interdisciplinary (Physics, Mechanics, Optics, Electronics, and Computer Science) AI problems are open ended, uncertainty dominated and use dynamic data.
10
How do humans and animals think and act? Cognitive Psychology and Cognitive Science Behaviorism
11
How does the Brain Work? Compare with Computer? 10**11 neurons vs 1 CPU (10**8 gates) 10 **-3 sec vs 10 **-10 sec 10**14 bits/sec vs 10**10 bits/sec Moores Law (doubles every 1.5 years) CPU gate count will equal neurons in 2020. Does this mean anything?
12
AI requires knowledge. A characteristic of intelligent people is that they possess more knowledge. Knowledge includes and requires data and information. Knowledge is a collection of related facts, procedures, models and heuristics that can be used in problem solving. Knowledge may be, specific, general, exact, fuzzy, (indistinct or blurred) procedural or declarative.
13
` `
Knowledge is provided by people who understand it. A physician treating a patient uses both knowledge and data. - The data is the patients record including symptoms, drugs given etc. - The knowledge is what the physician learnt in college, and with experience. Knowledge can be used in many situations. AI technique is a method to exploit knowledge. AI programs receive input from environment and give output to affect environment. The processing may involve common sense reasoning, expert reasoning, or logic programming.
14
` ` `
Knowledge representation is important because almost every computational intelligent activity depends on it to some degree Some KR schemes- First order Predicate Logic, Semantic networks, Frames, Scripts, Conceptual Dependencies Research in knowledge representation (KR) seeks to discover effective and efficient methods for representing information about all aspects of the world
15
16
Understand thinking. Make computers perform tasks that require intelligence if performed by people.
Asking if a computer can think is like asking if a submarine can swim. Dijkstra
17
Main Goal : Artificial intelligence. Sub Goal : Understanding human intelligence and how it is possible. Related subjects : Neurophysiology Cognitive Psychology (Artificial Neural Networks)
18
` ` `
To act humanly? (See Turing Test) To think humanly? (See Cognitive Science) Needs Human tests To think rationally? Logic and logicist tradition Game Theory and Rationality Act Rationally? (Back to Turing Test); limited rationality? Environment Important
19
1)
Use of computer metaphors has led to rich language for talking and thinking about thinking. Computer models force precision. Computer implementations quantify task requirements. Computer programs can be experimented on in ways that animal brains can not.
2)
3)
4)
20
Engineering : Solve real world problems using ideas about knowledge representation and handling. Scientific : Discover ideas about knowledge that helps explain various orts of intelligence.
21
Physical Symbol Hypothesis (Newell) : A physical symbol system has the necessary and sufficient means for intelligent action.
This hypothesis means that we can hope to implement this in the computer.
`
22
LEVEL 1 ISSUES Is it possible to create non human intelligence ? Can we model the brain ? Can we teach machines to think & solve problems like humans ?
23
What are our assumptions about intelligence ? What kind of techniques will be useful for solving AI problems ? At what level of detail are we trying to model human intelligence ?
24
Search - Avoids Combinatorial Explosion Perception - Involves Analog Signals With Noise Inference
25