Possible principles for aligned structure learning agents

Open in new window