Deepmind worked on reinforcement learning for plasma control back in 2022 and this also led to a paper in nature. I don't really understand the differences between their earlier work and this paper but deepmind don't seem to be involved in this one: https://deepmind.google/discover/blog/accelerating-fusion-sc...
The DeepMind paper in turn cites the authors of this paper, previously. One of the big differences in the current paper is that the experimental device is much larger and more powerful, and the duration of the shot is longer as well.