Start

Octi AI Agent

Octi AI Agent

2025

This agent utilizes a Monte Carlo Tree Search (MCTS) algorithm with a utility function to make optimal moves in the game of Octi. When choosing a move, it first checks for any immediate winning actions. If no such move is found, it creates a search tree by iterating through the four MCTS steps: selection, expansion, simulation, and backpropagation. The agent is available on GitHub and can be run locally.

Octi
Artifical Intelligence
Agent
MCTS
Utility Function
Zero-sum game
Screenshot of a website