It's not clear to me that you do get stronger guarantees because the setting and method is so similar to that of classical imitation learning. In both cases, we seek to learn a policy that is aligned with the expert (human). Supervised fine-tuning (behavioral cloning) is problematic because of distribution shift, i.e. the learned policy accumulates error (at a quadratic rate!) and visits states it did not see in training.
You say this failure mode is dangerous because of scheming AI and I say it's dangerous because the policy is OOD, but it appears you agre... (read more)
It's not clear to me that you do get stronger guarantees because the setting and method is so similar to that of classical imitation learning. In both cases, we seek to learn a policy that is aligned with the expert (human). Supervised fine-tuning (behavioral cloning) is problematic because of distribution shift, i.e. the learned policy accumulates error (at a quadratic rate!) and visits states it did not see in training.
You say this failure mode is dangerous because of scheming AI and I say it's dangerous because the policy is OOD, but it appears you agre... (read more)