Abstract
Stratego programming is still in its infancy. Although the best programs win occasionally against average human players, they have little chance against experts. One problem is that most Stratego programs have been modelled after perfect-information games such as chess. This note explains why Stratego is poorly suited for techniques developed for perfect-information games, and suggests a new direction based on information management that promises to produce superior results.
Get full access to this article
View all access options for this article.
