Flappy Bird AI using Evolution Strategies
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Mahdi Dibaiee 3819ce72c0 update web link 11 months ago
assets initial commit 6 years ago
demo initial commit 6 years ago
.gitignore web version 6 years ago
README.md update web link 11 months ago
demo.py initial commit 6 years ago
draw_chart.py feat(charts): draw reward charts 4 years ago
es.py feat(charts): draw reward charts 4 years ago
fig-log.jpg feat(charts): draw reward charts 4 years ago
game.py web version 6 years ago
load.npy fix: use a random seed to increase reproducibility 6 years ago
long.npy fix: increase space between bird and wall 6 years ago
package-lock.json feat(charts): draw reward charts 4 years ago
package.json feat(charts): draw reward charts 4 years ago
play.py initial commit 6 years ago
record.py fix: no need for gtk 6 years ago
requirements feat(charts): draw reward charts 4 years ago
train.py feat(charts): draw reward charts 4 years ago
win.py fix: increase space between bird and wall 6 years ago

README.md

Playing Flappy Bird using Evolution Strategies

After reading Evolution Strategies as a Scalable Alternative to Reinforcement Learning, I wanted to experiment something using Evolution Strategies, and Flappy Bird has always been one of my favorites when it comes to Game experiments. A simple yet challenging game.

The model learns to play very well after 3000 epochs, but not completely flawless and it rarely loses in difficult cases (high difference between two wall entrances). Training process is pretty fast as there is no backpropagation, and is not very costy in terms of memory as there is no need to record actions as in policy gradients.

Here is a demonstration of the model after 3000 epochs (~5 minutes on an Intel(R) Core(TM) i7-4770HQ CPU @ 2.20GHz):

after training

Before training:

Before training

There is also a a web version available for ease of access.

For each frame the bird stays alive, +0.1 score is given to him. For each wall he passes, +10 score is given.

Demonstration of rewards for individuals and the mean reward over time (y axis is logarithmic): reward chart

Try it yourself

You need python3.5 and pip for installing and running the code.

First, install dependencies (you might want to create a virtualenv):

pip install -r requirements

The pretrained parameters are in a file named load.npy and will be loaded when you run train.py or demo.py.

train.py will train the model, saving the parameters to saves/<TIMESTAMP>/save-<ITERATION>.

demo.py shows the game in a GTK window so you can see how the AI actually plays (like the GIF above).

play.py if you feel like playing the game yourself, space: jump, once lost, press enter to play again. 😁

pro tip: reach 100 score and you will become THUG FOR LIFE 🚬

Notes

It seems training past a maximum point reduces performance, learning rate decay might help with that. My interpretation is that after finding a local maximum for accumulated reward and being able to receive high rewards, the updates become pretty large and will pull the model too much to sides, thus the model will enter a state of oscillation.

To try it yourself, there is a long.npy file, rename it to load.npy (backup load.npy before doing so) and run demo.py, you will see the bird failing more often than not. long.py was trained for only 100 more epochs than load.npy.