How to prevent learning in reinforcement learning?

Question

With the dangers of AI lurking around the corner, I am wondering how much research has been done in preventing AI algorithms from learning and how to intentially confuse or disrupt them. For example, what can be done to a data stream to prevent reinforcement learning from learning?

mindcrime · Accepted Answer

It's not specific to reinforcement learning, but you might find this of interest:https://en.wikipedia.org/wiki/Adversarial_machine_learning

b20000 · Answer

came across this interesting paper - https://onlinelibrary.wiley.com/doi/10.1002/sat.1418?af=R