COS3751
Assignment 2
2023
Due date: 28 July 2023
UnisaGuides
, Preview…
Question 1
1.1. A utility function is a numerical measure of how good a game state is for a player. It is used in adversarial searches to
evaluate terminal states and guide the decision-making process towards the most advantageous moves(used to find
the best available move for a player.)
1.2. No. Agents can still get the ideal strategy when playing without all the information. This means the ideal strategy
includes the notion of limited information, however, a agent that has perfect information will most likely outperform
an agent that has limited/imperfect information for the same problem.
1.3. Forward pruning means that some nodes are pruned without even considering them.
Beam search [others are also considered] only considers a sample of the best moves at each ply, but this may lead to
the best move being pruned away.
1.4. Yes, the order in which nodes are examined in Minimax can have an impact on the efficiency of the search, but it does
not affect the correctness of the algorithm. The order in which nodes are examined in Minimax becomes significant
only when pruning techniques are applied. Otherwise minimax is an exhaustive search (for the parameters such as
depth – plys – it is configured for) and node orders do not matter.
In the absence of pruning, Minimax functions as an exhaustive search, exploring all possible moves up to a specified
depth (plys) without favoring any particular node order.
Question 2
2.1
Minimax, alpha/beta
A = -7,
Assignment 2
2023
Due date: 28 July 2023
UnisaGuides
, Preview…
Question 1
1.1. A utility function is a numerical measure of how good a game state is for a player. It is used in adversarial searches to
evaluate terminal states and guide the decision-making process towards the most advantageous moves(used to find
the best available move for a player.)
1.2. No. Agents can still get the ideal strategy when playing without all the information. This means the ideal strategy
includes the notion of limited information, however, a agent that has perfect information will most likely outperform
an agent that has limited/imperfect information for the same problem.
1.3. Forward pruning means that some nodes are pruned without even considering them.
Beam search [others are also considered] only considers a sample of the best moves at each ply, but this may lead to
the best move being pruned away.
1.4. Yes, the order in which nodes are examined in Minimax can have an impact on the efficiency of the search, but it does
not affect the correctness of the algorithm. The order in which nodes are examined in Minimax becomes significant
only when pruning techniques are applied. Otherwise minimax is an exhaustive search (for the parameters such as
depth – plys – it is configured for) and node orders do not matter.
In the absence of pruning, Minimax functions as an exhaustive search, exploring all possible moves up to a specified
depth (plys) without favoring any particular node order.
Question 2
2.1
Minimax, alpha/beta
A = -7,
