Perceptron Based Decision Making Unity DOTS
Forging Station
Apr 4, 2021 14:18
Create an account to hide videos, mark them watched, or save them to Don't Miss.
Create an account to hide videos, mark them watched, or save them to Don't Miss.
Create an account to hide videos, mark them watched, or save them to Don't Miss.
In this video, I try and explore the possibilities of having a single perceptron make sensible intelligent decisions. We have a system where a bot wanders around to consume food of type bread or venom to stay alive as long as possible. Consuming bread will add health whereas consuming venom removes some health but provides a speed boost.
It is up to the perceptron to figure the right bias values to switch at the right time to maximize the survivability of the bots!
There is a facility to limit the number of generations to omit those bots which show no evolution. This can happen if other bots have matured too fast which usually leads to scarcity of available resources.
Availability of consumables/resources greatly influences the overall system. In my case, I had to spawn bread every 0.5 seconds and venom every 1 second for 10 bots to evolve.
About Perceptron -
https://www.youtube.com/watch?v=ntKn5TPHHAk
GitHub -
https://github.com/ForgingStation/Perceptron-Based-Decision-Making
Gaming
#Forging Station
#Sumera Sankruthyayana
#Perceptron based decision making
#Perceptron
#Unity DOTS
#Unity ECS
#Decision
#Artificial Intelligence
#Layering
#Nodes
#Machine Learning
#Basic
#Single Perceptron
#Evolution
#Evolutionary Programming
#Evolved Code
#Perceptron based evolution