A

Agent — An autonomous entity that perceives its environment and takes actions to achieve goals.

Adversarial Training — A technique where two models compete (e.g., GANs) to improve each other.

B

Behaviour Tree — A hierarchical model for controlling NPC decision-making, widely used in game AI.

BOIDS — A flocking simulation algorithm for modelling group movement of characters or creatures.

G

Generative AI — AI systems that create new content (images, text, audio, 3D models) from training data.

GAN — Generative Adversarial Network. A framework where generator and discriminator compete to produce realistic outputs.

L

LLM (Large Language Model) — An AI model trained on vast text data capable of generating and reasoning about text. Examples: GPT-4, Claude, Gemini.

N

NavMesh — A data structure defining walkable surfaces for AI pathfinding.

NPC (Non-Player Character) — A game character controlled by AI rather than a human player.

P

PCG (Procedural Content Generation) — Algorithmic creation of game content such as levels, maps, items, and quests.

R

Reinforcement Learning (RL) — A ML approach where an agent learns through trial, feedback, and reward signals.

U

Utility AI — A decision-making system where NPCs score and choose actions based on weighted utility functions.

More terms added regularly. Suggest a term →