Hence, if we consider an user/warrior to be the agent who wants to reach to the treasure, then in each state on its path would be evaluated using rewards. It works using the concept of agent who wants to perform a task, reward system to help the agent evaluate different scenarios, an environment where the tasks needed to be performed and different states to make transition/take an action. Reinforcement learning is the science of making optimal decisions. Usage of Reinforcement learning on treasure hunt: Commit your source code, documentation and other supporting files to the git repository in GitHub.Build one or more reinforcement policies to model situational assessments, actions and rewards programmatically.Formulate ideas on how reinforcement learning can be used to find treasure efficiently while avoiding obstacles and opponents.Choose a programming language (Python, C/C++, Java).Think up a map-like environment with treasure, obstacles and opponents.Set up a new git repository in your GitHub account.We do the treasure hunting and monster fighting for you EECS 738 Project - The goal of this project is to implement a treasure hunt game to find the optimal path to treasure using Reinforcement learning.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |