Roland Pihlakas comments on A brief review of the reasons multi-objective RL could be important in AI Safety Research

[Roland Pihlakas]

The paper is now published with open access here:

https://​​link.springer.com/​​article/​​10.1007/​​s10458-022-09586-2