Zobrazeno 1 - 2
of 2
pro vyhledávání: '"Slaoui, Reda Bahi"'
Producing agents that can generalize to a wide range of visually different environments is a significant challenge in reinforcement learning. One method for overcoming this issue is visual domain randomization, whereby at the start of each training e
Externí odkaz:
http://arxiv.org/abs/1910.10537
Autor:
Clements, William R., Van Delft, Bastien, Robaglia, Benoît-Marie, Slaoui, Reda Bahi, Toth, Sébastien
Reinforcement learning agents are faced with two types of uncertainty. Epistemic uncertainty stems from limited data and is useful for exploration, whereas aleatoric uncertainty arises from stochastic environments and must be accounted for in risk-se
Externí odkaz:
http://arxiv.org/abs/1905.09638