Yes, the concept of _dreaming_ or _imagining_ has already been explored in reinforcement learning. For example, have a look at [Metacontrol for Adaptive Imagination-Based Optimization][1] (2017) by Jessica B. Hamrick et al., which is a paper that I gave a talk/presentation on 1-2 years ago (though I don't remember well the details anymore). There is also a blog post about the topic [Agents that imagine and plan][3] (2017) by DeepMind, which discusses two more recent papers and also mentions Hamrick's paper. In 2018, another related and interesting paper was also presented at NIPS, i.e. [World Models][2], by Ha and Schmidhuber. If you search for "imagination/dreaming in reinforcement learning" on the web, you will find more papers and articles about this interesting topic. [1]: https://arxiv.org/pdf/1705.02670.pdf [2]: https://worldmodels.github.io/ [3]: https://deepmind.com/blog/article/agents-imagine-and-plan