L5: Expert Systems, Heuristics, Second Boom Flashcards
What WAS weak methods in AI?
Early AI used generic methods intended to be applicable to wide range of problems.
* Have difficulty scaling because don’t incorporate problem-specific knowledge.
Were dubbed weak methods, since they only use knowledge that is weakly applicable
to the specific problem.
* Confusingly, nowadays “weak AI” indicates problem-specific AI programs, which are the converse of “strong AI”, which is used to indicate a (so far) hypothetical
generic problem solving program.
During 1970’s and 1980’s there was a shift towards problem-specific solutions.
What is an expert system?
An expert system is a computer system that emulates decision-making process of domain expert.
* Requires encoding a great deal of domain-specific knowledge.
* Reasoning often encoded using vast knowledge base of rules.
* Process often augmented by heuristics.
What is a heuristic?
Heuristics are criteria, methods, or principles for deciding which among several alternative courses of action promises to be the most effective in order to achieve some goal.
They are
* rules of thumb;
* not always applicable or optimal;
* methods for quickly finding something “good enough”;
* a compromise between simplicity, efficiency, accuracy, and completeness.
What is a heuristic?
Heuristics are criteria, methods, or principles for deciding which among several alternative courses of action promises to be the most effective in order to achieve some goal.
They are
* rules of thumb;
* not always applicable or optimal;
* methods for quickly finding something “good enough”;
* a compromise between simplicity, efficiency, accuracy, and completeness.
What is the nearest neighbour heuristic for the travelling salesman?
The nearest neighbour heuristic works as follows:
1. Start with an arbitrary node;
2. Choose as next node an unvisited node with minimum distance to current node;
3. Repeat until all nodes are visited, then return to starting node along shortest edge.
Notes:
* Solution depends on choice of initial city
* No general bounds on approximation factor
* Exist problem instances where this gives the worst possible solution.
What is alpha-beta pruning?
- Can speed up the code by not traversing
unnecessary nodes in the tree, while still returning the correct best move - α represents the minimum score that the maximising player is assured of based on the part of the search tree that has already been considered (-inf at start).
- β, on the other hand, at the same point
in the tree, represents the maximum score that the minimising player is guaranteed (inf at start).
When α ≥ β the remaining nodes and/or branches are pruned from the tree. This is because the relevant player already knows they can achieve better results from moves that have already been seen.
What was the fifth generation computer systems project?
Fifth generation computer systems project initiated in Japan in 1982.
- Gen. 5 was to include massive parallelization and logic programming
The fifth generations computers were to…
* use super large scale integrated chips;
* have artificial intelligence (sic);
* be able to recognize images and graphs;
* be able to solve complex problems, e.g. decision making, logical reasoning;
* work with natural language.
