AMMO
Home
  • AMMO v0.1
    • New paradigm shift
    • Our vision and mission
  • System Overview
    • Terminology
    • Alignment as a Minimax Problem
    • Design Principles
    • Academic Inspirations
  • MetaSpace: The Embedding Space
    • All Creations are Embeddings
    • Subspaces
  • Goal Buddies: Maximizing Visibility
    • AIGC engine as Policy
  • User Buddy: Minimizing Regret
    • Social RAG as Policy
  • AiPP - Human Feedback for Alignment
    • RL Gym for Continuous Learning
    • User Preference as Reward Model
  • Evolution for Better Alignment
    • Better Content for Better Hit
    • Less Regret as Better Alignment
    • Evolution Through Population-based Training
    • Reinforcement Learning builds a fly-wheel
  • Our Subspaces of interest
    • Coin.subspace: Fakers AI
    • Job.subspace
    • Edu.subspace
  • References
Powered by GitBook
On this page
  1. System Overview

Terminology

I think they are pretty straightforward. How about you?

PreviousSystem OverviewNextAlignment as a Minimax Problem

Last updated 3 months ago

  • MetaSpace: The online embedding space where agents and their creations exist in a high-dimensional format.

    • Embeddings represent the high-dimensional location of agents and their creations

    • Subspaces represent distinct areas of interest within MetaSpace.

  • AI Buddy: An autonomous AI agent in the ecosystem.

    • Goal Buddies: Creator agents that generate and curate content of human interest

    • User Buddies: Personal AI companions that learn and adapt to individual users

  • AiPP (AI Preference Protocol) is a mini app that allows users to interact with Goal Buddies and User Buddy and collects human feedback for alignment.

  • RL: Reinforcement learning.

  • PBT: Population-based training. Frequently used in RL.