Faur/Discrete-World-Models

Code for implementing discrete probalbistic world models for applications within Reinforcement learning.

Discrete World models

[Work in progress]

Code for implementing discrete probablistic world models for applications within reinforcement learning. Work is particularly inspired by:

• Ha, D., & Schmidhuber, J. (2018). World Models. https://doi.org/10.5281/zenodo.1207631
• Maddison, C. J., Mnih, A., & Teh, Y. W. (2016). The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables. Retrieved from http://www.stats.ox.ac.uk/~cmaddis/pubs/concrete.pdf
• Jang, E., Gu, S., & Poole, B. (2016). Categorical Reparameterization with Gumbel-Softmax. Retrieved from http://arxiv.org/abs/1611.01144
• Oord, A. van den, Vinyals, O., & Kavukcuoglu, K. (2017). Neural Discrete Representation Learning. Retrieved from https://arxiv.org/pdf/1711.00937.pdf