References
- Rich, Knight and NAIR
- Patterson
- Janakiraman
Coverage
- Introduction
- State-space search
- Game playing
- Predicate logic
- Other logics
- Structured knowledge representations
- Expert systems
- Planning
- Natural language processing
- Learning
introduction
Contents
- Definition
- History
- Applications
- AI technique
- Some problems
Definition
- Study of how to make computers do things at which, at the moment, people are better
- Examples:
- Perception, Natural language understanding …
- The computer is made better at some tasks such as:
- A game playing, Computing, Theorem proving, Expert problem solving …
History
- Began around 1960
- Logic Theorist – Novell 1963
- Was able to prove several theorems from the first chapter of Russell’s Principia (1950)
- Game playing and theorem proving was considered as a sign of intelligence – involves thinking and intelligent planning
- Computers do it by their inherent power of searching quickly the innumerable solution paths
- The search process is not intelligent act and can be easily programmed
- Samuel’s checker playing program released in 1960s
- This was able to learn and become more proficient as more games are played
- GPS built by Newell, Shaw and Simon in 1963
- Could do simple routine tasks as we do when we go to the office every day
- Solved using logic expressions
- Not a lot of knowledge built-in
- Some natural language processing programs
- ELIZA (Weizenbaum 1966) PARRY (COLBY 1975)
- Superficial understanding
- STUDENT (Bobrow 1968)
- Could solve high school word problems
- SHRDLU (Winogard 1973)
- All these operate on simplified worlds
- Expert systems
- MYCIN, PROSPECTUS, DENDRAL …
- Solve problems (make decisions like an expert) in a specific domain
