The horizon effect is a problem in artificial intelligence where, in many games, the number of possible states or positions is immense and computers can only feasibly search a small portion of it, typically a few ply down the game tree. Thus, for a computer searching only five ply, there is a possibility that it will make a move which is detrimental, but the detrimental effect is not visible because it does not search to the depth of the error (i.e. beyond its horizon).
When evaluating a large game tree using techniques such as minimax or alpha-beta pruning, search depth is limited for feasibility reasons. However, evaluating a partial tree may give a misleading result. When a significant change exists just over the 'horizon' of the search depth, the computational device falls victim to the horizon effect.
The horizon effect can be mitigated by extending the search algorithm with a quiescence search. This gives the search algorithm ability to look beyond its horizon for a certain class of moves of major importance to the game state, such as captures.
Rewriting the evaluation function for leaf nodes and/or analyzing sufficiently more nodes will solve many horizon effect problems.
Read more about Horizon Effect: Example
Famous quotes containing the words horizon and/or effect:
“However closely people are attached to one another, their mutual horizon nonetheless includes all four compass directions, and now and again they notice it.”
—Friedrich Nietzsche (1844–1900)
“Considered physiologically, everything ugly weakens and saddens man. It reminds him of decay, danger, impotence; it actually reduces his strength. The effect of ugliness can be measured with a dynamometer. Whenever anyone feels depressed, he senses the proximity of something “ugly.” His feeling of power, his will to power, his courage, his pride—they decline with ugliness, they rise with beauty.”
—Friedrich Nietzsche (1844–1900)