Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.
  2. Join us on March 30, 2023, between 5 am & 1 pm EST, in the Performance Profiling Dev Blitz Day 2023 - Q&A forum and Discord where you can connect with our teams behind the Memory and CPU Profilers.
    Dismiss Notice

Decision and action structure with ML agents (help needed)

Discussion in 'ML-Agents' started by GoRoMo, Sep 11, 2022.

  1. GoRoMo


    Mar 7, 2021
    Hello, I'm really new to ML agents and I don't know how it works very well.

    After having some practice with some examples I decided to build one of my first projects using ML agents.

    I want the IA to do two things, decide which actions to perform, and to realize the action it decided to.
    As they will be multiple actions I want to do, I thought the best option would be to train every action as different IA's, and then train a central IA to decide which actions it should perform in different situations.

    The concept itself is simple, but I can't even think on how to implement it.
    I'll be trying to build a system that works like that, but I'm doubtful it may even be doable with ML agents, but I can't say nothing useful as I know almost nothing about ML agents.

    I would thank any possible help, in building the system or finding alternatives.
    Thanks in advance.