Skip to content

Latest commit

 

History

History
16 lines (16 loc) · 648 Bytes

README.md

File metadata and controls

16 lines (16 loc) · 648 Bytes

Flappy-Birds

NN + RL for Flappy Birds

An agent trained to play Flappy Birds using Q learning and Neural Networks. Our project for the Neural Networks course in 2019. The learning environment used is from PyGame-Learning-Environment.

demo

Results of our best agent ( vertical -> score, horizontal -> number of games played):
demo
Two hidden layers, with leaky relu and tanh as activation functions and Adadelta as optimizer.