Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Machine learning in the “Settlers of Catan” strategic board game

Papadam Diamantis-Rafail

Full record

---

Summary

Despite recent deep neural network superhuman performance in many strategic board games, such as Chess and Go, there does not yet exist an algorithm that beats “Settlers of Catan” professional human players. Towards this direction, we present a combination of modern machine learning with traditional tree-based adversarial search algorithms and achieve performance close to the state-of-the-art in initial settlement placement. In particular, we use a generalization of the classic Minimax search algorithm, known as Max^n , with the novelty that the evaluation function at the leaf nodes is the result of a forward pass in a trained convolutional neural network. Our work consists of two distinct parts that can work independently. The first is the Max^n algorithm implementation that could use any evaluation function. The second is the neural network, which acts as an evaluation function and could be plugged into any adversarial search algorithm. After 10000 simulated games, which is a sufficient number for the demanding strategic board game “Settler of Catan”, we achieve performance close to the state-of-the-art; with the advantage that, in contrast to the state-of-the-art one, our approach’s runtime is acceptable by human players.

Available Files

Services