*Goodhart
Thanks! Corrected (though it is indeed a good hard problem).
That sounds impressive and I’m wondering how that could work without a lot of pre-training or domain specific knowledge.
Pre-training and domain specific knowledge are not needed.
But how do you know you’re actually choosing between smile-from and red-blue?
Run them on examples such as frown-with-red-bar and smile-with-blue-bar.
Also, this method seems superficially related to CIRL. How does it avoid the associated problems?
Which problems are you thinking of?
That sounds like a black-box approach.
Human’s not knowing what goals we want AI to have and the riggability of the reward learning process. Which you stated were problems for CIRL in 2020.
Thanks! Corrected (though it is indeed a good hard problem).
Pre-training and domain specific knowledge are not needed.
Run them on examples such as frown-with-red-bar and smile-with-blue-bar.
Which problems are you thinking of?
That sounds like a black-box approach.
Human’s not knowing what goals we want AI to have and the riggability of the reward learning process. Which you stated were problems for CIRL in 2020.