Karteikarten
SCC 361 - Artificial Intelligence
Karten in diesem Stapel (21)
- 1What is State Space?The set of all states that are reachable from an initial state by any sequence of actions
- 2What is a Path in state diagrams?A sequence of actions leading from one state to another
- 3How do we measure the performance of a search?1. Completeness
2. Time complexity
3. Space Complexity
4. Optimality - 4What does Completeness refer to?Is the strategy guaranteed to find a solution when one exists?
- 5What does Time Complexity refer to?How long does it take to find a solution
- 6What does Space Complexity refer to?how much memory does it require to perform the search
- 7What does optimality refer to?Does the strategy find the best-quality solution when more then one solutions exists
- 8When would A* Search not find a solution?1. Node with an infinite branching factor
2. A path with a finite path cost but an infinite number of nodes - 9When is A* Search complete?1. There is a finite branching factor
2. Every operator costs at least some positive (epsilon) - 10How to decide which attribute to split on?Idea: a good attribute to choose, is one that reduces uncertainty
e.g. split a data set into subsets that are all positive or all nagative - 11What is Probability Theory?Describes what an agent should believe, on the basis of evidence
- 12What is Utility TheoryDescribes what an agent wants
- 13What is Decision Theory?Puts the Probability Theory and Utility Theory together, to describe what an agent should do next
- 14What is Elitism?In GA's, Elitism copies the best chromosome to the new population
- increases performance since it prevents losing the best found solution - 15What are two replacement strategies in GA?1. Steady-State Selection
2. General Selection - 16What is Steady-State Selection?○ In every generation, select a few high fitness chromosomes for creating new offspring
○ Remove some low fitness chromosomes
○ Replace these with new offspring
Remainder of population remains - 17What is General Selection?All population members are removed on each generation
- 18Parameter's of a GA- population size
- encoding choices
- crossover probability
- mutation probability
- replacement strategy
- fitness function - 19Advantages to GA?- Parallelism
- Less likely to get stuck in local extrema than some other optimisation methods
- Relatively easy to implement - 20Disadvantages- Choosing an encoding and fitness function can be difficult
- computational time
- no guarantee on a solution
- strong dependence on parameters - 21How to model games that include chance?treat the randomization as another player
Informationen
- Beschreibung
- Karteikarten am SCC 361 - Artificial Intelligence, erstellt von Lewis Stones am 12/04/2018.
- Erstellt von
- Lewis Stones
- Aufrufe
- 1
- Bewertung