Question 46
Pruning occurs in $\Alpha-\Beta search whenever$ which condition is met?
When $\alpha >= \beta$, the current player knows the opponent can force a worse outcome elsewhere, so further exploration of this branch is pruned.
Pruning occurs in $\Alpha-\Beta search whenever$ which condition is met?
When $\alpha >= \beta$, the current player knows the opponent can force a worse outcome elsewhere, so further exploration of this branch is pruned.
When full depth search is impossible in real-time games, Minimax replaces terminal utility functions with:
Games like Chess use heuristic evaluation functions at cutoff depths to estimate board strength when full tree traversal is impossible.
Backward Chaining in expert systems is considered to be a:
Backward Chaining starts with a hypothesis or goal state and works backward to find supporting evidence or facts in the knowledge base.
The Horizon Effect in game-playing programs refers to:
The horizon effect occurs when an agent stalls an inevitable serious damage move by taking stalling actions that push the threat past the fixed search horizon.
Monte Carlo Tree Search (MCTS) relies on which four core phases during game tree expansion?
MCTS (used in AlphaGo) executes 4 steps iteratively: Selection (UCB1), Expansion, Random Simulation/Rollout, and Backpropagation of results.