An open API service indexing awesome lists of open source software.

https://github.com/ali2210/super-duper-broccoli

Strategies for implement AI
https://github.com/ali2210/super-duper-broccoli

ai description gametheory

Last synced: 2 months ago
JSON representation

Strategies for implement AI

Awesome Lists containing this project

README

        

# super-duper-broccoli
Strategies for implement AI

## Mathematical Model that solve games
1. Game Theory
2. Neural Nets
1. Binary Labeling (Easy but not reality engage)
2. Multiclass labeling (Easy but not reality engage)
3. Probalitics / Decision (ok purely mathematical)
4. Logics (symbolic annotation)
5. Re-enforcement Learning (Reality and limited unless agent have episodic memory)

## Game Theory
https://www.youtube.com/watch?v=DXH2EGjRpY4
In Game theory imagine you have two player Falcon and snake. Eagle want to kill snake and vice versa. While Imagine a suitation where both players
bound with rules of play. According to rule of play Eagle grab the snake and take upto > 15000 ft. Which may cause remain unconsciousness and lost
life battle.While Split venom at 1-2 meter. Otherwise serious injury or may be cause of dead. Who win nobody knows.

-----------------------------------------------------------------------------------------------
Eagle (win) Snake(win)

Snake(lose) Hawk Altidude >= 15000ft Snake split venom >= 0 - 2 meter Eagle(lose)

Snake(win) Snake split venom > 2 meter Eagle(win)
and Eagle altuide <= 10000 ft
---------------------------------------------------------------------------------------------------

As you know If Eagle want to win then either death of snake or choose Snake and Eagle win and same with snake.
The Real Challenge how long this match you can watch. Problem calculate equation either through probabilistic
methods / algorithms or through neural nets weight, Fuzzy Logic. Interesting is this unless player die or quit
this match remain continue. Equation Return the best probablity of an event which is unknow & intact players
with this probability as experience.