@article{therelationshipbetweendynamicprogramming,
  title = {Reward Maximisation through Discrete Active Inference},
  author = {Lancelot Da Costa and Noor Sajid and Thomas Parr and Karl Friston and Ryan Smith},
  year = {2020},
  eprint = {2009.08111},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2009.08111v4},
}