Week 4 Flashcards

1
Q

Minimax search

A

Minimax search is depth first search on the game tree.
1. The graph is built in real time
2. The goal is to evaluate the children of the current node to find the best node

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

When using mini-max, should you load the entire game tree?

A

No, this puts too much of a limit on the size of trees which can be searched.

Use depth-first search instead

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How can we remove the assumption of two-players from minimax?

A

By treating all other players as a single MIN player.
- No longer produces an equilibrium
- Assumes the opponents are working together to gang up on Player 1
- Still produces worst-case analysis for layer 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Why does treating all opponents in minimax as one player not produce an equilibria?

A

Suppose you are player 1, and player 2 has two possible moves:
1. Move A: a move which causes player 3 to win
2. Move B: a move which continues the game

Player 2 will always make move B

Minimax (Assuming Player 2 and 3 work together), assumes player 2 will make move A

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How can we remove the assumption of no chance from the minimax algorithm

A

Treat the value of a chance node as the expectation of all its children. Sometimes called “expectimax”

Uses a method called “Counterfactual Regret Minimization”, beyond the scope of this course

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Win-loss solve

A

Value of a MAX node:
- WIN if any of it’s children are WIN nodes
- LOSS if all of its children are LOSS nodes

Value of a MIN node:
- WIN if all of it’s children are WIN nodes
- LOSS if any of its children are LOSS nodes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Why can win-loss solve be much faster than minimax?

A

It doesn’t have to search the entire tree

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is alpha-beta pruning

A

Each node contains a range [alpha,beta] where:
- alpha is the maximum lower bound of the value of the node
- beta is the minimum upper bound of the value of the node
- where beta <= alpha prune the subtree containing that node
- Start with the range [-inf,inf]
- Only MAX nodes can change alpha; only MIN nodes change beta
- The range is passed down the tree unchanged
- A child can change the range of its parent

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What do we do for large games using minimax

A

Use an “evaluation function” (“heuristic”) to approximate the value of the deepest nodes we can reach, if they are not terminal nodes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What would be a good heuristic for tic-tac-toe?

A

Number of lines where X could win - number of lines where O could win

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How to find good heuristics for minimax?

A

Admissible and monotonic does not apply.

  • Requires knowledge of the game
  • Need to make sure the heuristic is correctly normalised against values returned at actual terminal nodes

Often trial and error is used:
- Player them against each other in multiple games (from both starting positions)
- Play them both against other heuristics
- The one that wins the most is likely better

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How to find the best weights for a linear combination of a set of k heuristics?

A

Hand tweaking
- Often done by guess work

Stochastic hillclimber
- Set random weights with a zero-mean over and over and find the combination that evaluates the best

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the standard way to play two-player, perfect information games?

A

Use a good mini-max tree search algorithm with effective pruning.

Use effective heuristic evaluation function

Major issue: Finding good heuristic evaluation functions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are some possible extensions to the minimax algorithm

A
  • Database of opening moves
  • Database of end games
  • Move reordering to increase pruning
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What game did minimax fail to win?

A

Go

How well did you know this?
1
Not at all
2
3
4
5
Perfectly