IAI Chapter 4 Games
Questions about the lecture 'Introduction to Artificial Intelligence' of the RWTH Aachen Chapter 4 Games
Questions about the lecture 'Introduction to Artificial Intelligence' of the RWTH Aachen Chapter 4 Games
-
- 1 / 10
-
Lernkarten
What can be done to limit the game tree ? [3]
1. Early pruning with replacing terminal-test by cut-off-test in quiescent states
2. Good evaluation function with replacing utility by evaluate
3. Horizon problem due to fixed depth search
What are the characteristics of 2-person games? [3]
1. Players are MAX and MIN from MAX points of view
2. Initial board position and assignment MAX -inf and MIN inf // I will lose and opponent will win
3. Utility function to determine (quality of) outcome // Win, lost or draw
What is the definition of minimax? [6]
1. Generate complete game tree
2. Apply utility function for each terminal state
3. Assign minimum value to MIN parent node
4. Assign maximum value to MAX parent node
5. Choose child with maximum utility at root level
6. Assume that opponent chooses always the best move
What is the definition of alpha beta? [3]
1. Generate game tree to certain depth
2. Compute utility/evaluation of nodes
3. Do not continue expanding nodes which are already worse // Do not waste to see how bad it is
Since when have games been studied?
1950s
What are games regarded to general search? [2]
1. Special variant
2. With only accessible states
What are the actions of games?
Moves of player // Complication for more than player
What kind of problems are usually described by games?
Contingency problems due to uncertain moves of opponent // Usually maximaizing damage
What is a problem of games?
Usually unsolveable due to number of possible sequences
What is added to tree if games involves chance?
An additional layer
