NEAT Tetris

Learns to play Tetris on its own! 

Found at: this link 

WHAT IS NEAT TETRIS?

Our project is an agent which has been trained to play the game of Tetris. This system uses a genetic algorithm known as NEAT, based on the process of natural selection, to refine the A.I.'s Tetris gameplay strategy. 
A population of A.I.'s each with slightly different strategies, will be created from randomly generated neural network configurations, and each play short games, untill all are finished 
Then, the two top-scoring networks("parents") will be bred together to make a new population, and the process repeats for several generations. 
And finally, the best A.I. in the last generation should be better than the one you started with. 

Genetic Algorithm

Takes the best parts of the agents and combines them together.

Neural Network

This is what makes the player play. It takes all the inputs, processes them and decides what move to make.

A SHORT SHOWCASE OF ANGELINA

Angelina effectively and efficiently solving your Tetris needs

A SHORT SHOWCASE OF BRAD V1

Brad not so effectively and efficiently solving your Tetris needs

A showcase of NEAT used for Mario

A short explaination of NEAT, giving a general idea of what it is and how it works

WHAT OUR AGENTS DO

“I give each player one chance to play and judge them based on how well they perform at the end.”

Brad V1

“I give each player three chances to play and judge them based on how well they perform at the end.”

Brad V2

“I give each player one chance to play and not only do I judge them based on how well they perform at the end but also on every signle move they make.”

Angelina V1

MEET THE TEAM!

Keniel Peart

Rajay Bitter

Kimberly Soares

Shanice Bryan