Adversarial Search

Question 1

Reinforcement learning techniques has eclipsed __________ search for some games

Answer 1

alpha-beta

Question 2

We represent 2-player games like chess as a ____ ______

Answer 2

game tree

Question 3

In the game tree, nodes alternate between _____

Answer 3

moves

Question 4

In a min/max search, MIN whats to ______ the objective function and MAX wants to ______ it

Answer 4

minimize, maximize

Question 5

The _________ algorithm is the basic search techniques that computers use for making moves

Answer 5

MINIMAX

Question 6

Alpha-beta pruning is a technique for efficiently _________ certain parts of the game tree

Answer 6

discarding

Question 7

AlphaZero learned how to play chess in less than ___ hours of _____, but it did needs tons of _________ power

Answer 7

10, self-play, computational

Question 8

AlphaZero uses a technique called…

Answer 8

Monte-Carlo Tree Search

Question 9

What is the basic idea of the Monte-Carlo Tree Search?

Answer 9

• Pick random successor states until the game is played out
• Repeat thousands of times
• Select the move leading to the most wins

Question 10

AlphaZero has also become a master at ______ and __

Answer 10

Shogi, Go

Question 11

_____ _______ was the first machine to beat a grandmaster in a tournament. It was later brought to IDB and renamed ____ _____

Answer 11

Deep Thought, Deep Blue