Reinforcement Learning Agents Navigate Turbulent Odor Plumes Without Spatial Maps

Reinforcement learning agents can learn to efficiently navigate to an odor source in a turbulent environment using only temporal features of odor cues, without any prior spatial information.