English

Making Deep Q-learning methods robust to time discretization

Machine Learning 2019-01-30 v2 Machine Learning

Abstract

Despite remarkable successes, Deep Reinforcement Learning (DRL) is not robust to hyperparameterization, implementation details, or small environment changes (Henderson et al. 2017, Zhang et al. 2018). Overcoming such sensitivity is key to making DRL applicable to real world problems. In this paper, we identify sensitivity to time discretization in near continuous-time environments as a critical factor; this covers, e.g., changing the number of frames per second, or the action frequency of the controller. Empirically, we find that Q-learning-based approaches such as Deep Q- learning (Mnih et al., 2015) and Deep Deterministic Policy Gradient (Lillicrap et al., 2015) collapse with small time steps. Formally, we prove that Q-learning does not exist in continuous time. We detail a principled way to build an off-policy RL algorithm that yields similar performances over a wide range of time discretizations, and confirm this robustness empirically.

Keywords

Cite

@article{arxiv.1901.09732,
  title  = {Making Deep Q-learning methods robust to time discretization},
  author = {Corentin Tallec and Léonard Blier and Yann Ollivier},
  journal= {arXiv preprint arXiv:1901.09732},
  year   = {2019}
}
R2 v1 2026-06-23T07:24:10.167Z