PyBrain

In order to make my research as reproducible as possible, most of the algorithms and benchmarks I use are available open-source through the PyBrain machine learning library.

PyBrain is written in Python, and while my contributions focus on (recurrent) neural networks, reinforcement learning and black-box optimization, it also contains algorithms for supervised learning, unsupervised learning, evolution, visualization, robotic simulations and more...

To get started, get it from github, read the tutorials, play around with the (40+) examples, and if you have more questions, contact the mailing list.

If you use PyBrain in your research, you can add your work to this list and cite our paper[16]:

@article{pybrain,
title = {{PyBrain}},
author = {Schaul, T. and Bayer, J. and Wierstra, D. and Sun, Y.
and Felder, M. and Sehnke, F. and R\"{u}ckstie\ss, T. and Schmidhuber, J.},
journal = {Journal of Machine Learning Research},
pages = {743--746},
volume = {11},
year = {2010}

}