Artificial Intelligence is a collection of hard problems that humans and other living things can solve, but for which we don’t have good algorithms for solving.

example understanding spoke the natural language, medical diagnosis, circuit design, learning, self-adaptation, reasoning, chess-playing, proving math theories, etc.

• Artificial Intelligence is a process of making a machine or a program that
• Learn and understand like a human.
• Acts like a human (Turing test).
• Thinks like a human (human-like patterns of thinking steps).
• Acts or thinks rationally (logically, correctly).

It is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to biologically observable methods.

• Artificial Intelligence is the study of how to make computers just like humans. That means how to make computers do things that people do better
• Some problems used to be thought of as AI but are now considered note. g., compiling Fortran(suited to numeric computation and scientific computing) in 1955,
• symbolic mathematics (manipulate mathematical equations) in 1965 proving math theories.
• Artificial Intelligence is the study and design of intelligent agents where an intelligent agent is a system that interacts with its environment and takes actions that maximize its chances of success.

INTELLIGENCE 

Intelligence is the computational part of the ability to achieve goals in the world. Varying kinds and degrees of intelligence occur in people, many animals, and some machines. Artificial Intelligence is the study of how to make computers make things which at the moment people do better. Examples: Speech recognition, Smell, Face, Object, Intuition, Inferencing, Learning new skills, Decision making, Abstract thinking.

History of Artificial Intelligence

• Artificial Intelligence has roots in several scientific disciplines
• computer science and engineering (hardware and software)
• philosophy (rules of reasoning)
• mathematics (logic, algorithms, optimization)
• cognitive science and psychology (modeling high-level human/animal thinking)
• neural science (model low level human/animal brain activity)
• linguistics

The birth of AI (1943 –1956)

Pitts and McCulloch (1943): a simplified mathematical model of neurons (resting/firing states) can realize all propositional logic primitives (can compute all Turing computable functions)

1. Allen Turing: Turing machine and Turing test (1950).
2. Claude Shannon: information theory; the possibility of chess-playing computers.

Early enthusiasm (1952 –1969)

1956 Dartmouth conferenceJohn McCarthy (Lisp); Marvin Minsky (first neural network machine); Alan Newell and Herbert Simon (GPS);

Emphasize on intelligent general problem solvingGSP (means-ends analysis); Lisp (Artificial Intelligence programming language); Resolution by John Robinson (the basis for automatic theorem proving); heuristic search (A*, AO*, game tree search)

Emphasis on knowledge (1966 –1974)

• domain-specific knowledge is the key to overcome existing difficulties
• knowledge representation (KR) paradigms
• declarative vs. procedural representation

Knowledge-based systems (1969 –1999)

• DENDRAL: the first knowledge-intensive system (determining 3D structures of complex chemical compounds)
• MYCIN: first rule-based expert system (containing 450 rules for diagnosing blood infectious diseases)EMYCIN: an ES shell
• PROSPECTOR: a first knowledge-based system that made a significant profit (geological ES for mineral deposits)

Artificial Intelligence became an industry (1980 –1989)

1. wide applications in various domains
2. commercially available tools
3. Current trends (1990 –present)
4. more realistic goals
5. more practical (application oriented)
6. distributed AI and intelligent software agents
7. the resurgence of neural networks and emergence of genetic algorithms

Programming languages for Artificial Intelligence

The programs for AI problems can be written with procedural languages like PASCAL or declaration languages like PROLOG. Generally, relational languages like PROLOG or LISP are preferred for symbolic reasoning in AI. However, if the program requires much arithmetic computation (say for the purpose of uncertainty management), then procedural languages would be preferred.

Recently some shells are available, where the user needs to submit knowledge only and the shall offer the implementation of both symbolic processing simultaneously.

Easy Problems in AI

It’s been easier to mechanize many of the high-level cognitive tasks we usually associate with “intelligence” in people

Example for - symbolic integration, proving theorems, playing chess, some aspect of medical diagnosis, etc.

Hard Problems in AI

It’s been very hard to mechanize tasks that animals can do easily catching prey and avoiding predators.

• interpreting complex sensory information.
• modeling the internal states of other animals from their behavior.
• working as a team.

Examples of AI problems

Expert Consulting Systems

  A key problem in the development of an Expert Consulting System is how to represent and use the knowledge that human experts in these subjects obviously possess and use. This problem is more difficult by the fact that the expert knowledge in any important field is imprecise and uncertain.

Theorem Proving

  Finding proof of a mathematical theorem requires the following intelligence. Requires the ability to make deductions from hypothesis.  

• It demands intuitive scales such as guessing which path should be proved first to help to prove the theorem.
• It also requires judgments to guess accurately about which previously proven theorems in a subject area will be useful in the present proof. 
• Also sometimes it is needed to break the main problem into sub-problems to work on independently.

Robotics

    It deals with the problems of controlling the physical actions of a Mobile Robot.

 Automatic Programming

    In automatic programming, a system takes in a high-level description of what program is to accomplish and produce a program.

 Perceptional Problems

Computers are made to see their surroundings by fitting T.V inputs.

 Also, they are made to hear speaking voices by providing microphone inputs.

 But it requires the processing of large base knowledge about the things being perceived.

 Natural Language Processor

This field is concerned with the efforts of making computers understand spoken and written languages. 

To understand sentences about a topic, it is necessary not only a lot about the vocabulary and grammar but also a good deal about the topic so that unstated assumptions can be recognized.

Task Domains of Artificial Intelligence

Mundane Tasks: 

1. Perception
2. Vision
3. Speech
4. Natural Languages
5. Understanding
6. Generation
7. Translation
8. Common sense reasoning
9. Robot Control
10. Formal Tasks
11. Games: chess, checkers, etc

Mathematics: Geometry, logic, Proving properties of programs

Expert Tasks:

Engineering ( Design, Fault finding, Manufacturing planning)

• Scientific Analysis
• Medical Diagnosis
• Financial Analysis

Think like humans

• Cognitive science approach
• Focus not just on behavior and I/O but also look at the reasoning process.
• The computational model should reflect “how” results were obtained.

GPS (General Problem Solver): Goal not just to produce human-like behavior (like ELIZA), but to produce a sequence of steps of the reasoning process that was similar to the steps followed by a person in solving the same task.

Law of thoughts

Develop formal models of knowledge representation, reasoning, learning, memory, problem-solving, that can result in algorithms. There is often an emphasis on provably correct systems, and guarantee finding an optimal solution.

For a given set of inputs, generate an

appropriate output that is not necessarily correct but gets the job done.

A heuristic(heuristic rule, heuristic method) is a rule of thumb, strategy, trick, simplification, or any other kind of device which drastically limits the search for solutions in large problem spaces. Heuristics do not guarantee optimal solutions; in fact, they do not guarantee any solution at all: All that can be said for a useful heuristic is that it offers solutions that are good enough most of the time.

Not interested in how you get results, just the similarity to what human results are.

Exemplified by the Turing Test (Alan Turing, 1950).

Acting Humanly: The Turing Test

Alan Turing (1912-1954)

“Computing Machinery and Intelligence” (1950)

Turing Test

• Three rooms contain a person, a computer, and an interrogator.
• The interrogator can communicate with the other two by teleprinter.
• The interrogator tries to determine which is the person and which is the machine.
• The machine tries to fool the interrogator into believing that it is the person.
• If the machine succeeds, then we conclude that the machine can think.

What would a computer need to pass the Turing test?

Natural language processing: to communicate with the examiner.

Knowledge representation: to store and retrieve information provided before or during interrogation.

Automated reasoning: to use the stored information to answer questions and to draw new conclusions.

Machine learning: to adapt to new circumstances and to detect and extrapolate patterns.

Vision (for Total Turing test): to recognize the examiner’s actions and various objects presented by the examiner.

Motor control (total test): to act upon objects as requested.

Other senses (total test): such as audition, smell, touch, etc.

Data

  Data are the facts and figures about a particular activity.

Data Processing

  The process of collecting all the required data together to produce meaningful information. 

Information

  Information is obtained by processing the data into meaningful form

Knowledge

  It is a structured representation of all the facts of an AI problem

Knowledge base

  The facts or assertions about the problem domain.

Database

  The storage medium for the state variables.

Eliza

 A program that simulated a psychotherapist interacting with a patient and successfully passed the Turing Test.

• Coded at MIT during 1964-1966 by Joel Weizenbaum.
• The first script was DOCTOR.
• The script was a simple collection of syntactic patterns not unlike regular expressions
• Each pattern had an associated reply which might include bits of the input (after simple transformations (my your).

The Loebner Contest

1. A modern version of the Turing Test, held annually, with a $100,000 cash prize.
2. http://www.loebner.net/Prizef/loebner-prize.html
3. Restricted topic (removed in 1995) and limited time.
4. Participants include a set of humans and a set of computers and a set of judges.
5. Scoring
6. Rank from least human to most human.
7. The highest median rank wins $2000.
8. If better than a human, win $100,000. (Nobody yet…)

What can AI systems do?

• Here are some example applications
• Computer vision: face recognition from a large set
• Robotics: autonomous (mostly) automobile
• Natural language processing: simple machine translation
• Expert systems: medical diagnosis in a narrow domain
• Spoken language systems:~1000 word continuous speech
• Planning and scheduling: Hubble Telescope experiments
• Learning: text categorization into ~1000 topics
• Games: Grand Master level in chess (world champion), checkers, etc.

What can’t AI systems do yet?

• Understand natural language robustly (e.g., read and understand articles in a newspaper)
• Surf the web
• Interpret an arbitrary visual scene
• Learn a natural language
• Construct plans in dynamic real-time domains
• Refocus attention in complex environments
• Perform life-long learning

AI Technique

1. Intelligence requires Knowledge
2. Knowledge possesses less desirable properties such as:
3. Voluminous
4. Hard to characterize accurately
5. Constantly changing
6. Differs from data that can be used
7. A technique is a method that exploits knowledge that should be represented in such a way that:
8. Knowledge captures generalization
9. It can be understood by people who must provide it
10. It can be easily modified to correct errors.
11. It can be used in a variety of situations.