Rohin Shah
TalkRL: The Reinforcement Learning Podcast
00:00
Recommenders and Recommendation Systems
i'm really concerned, like many people are, about the disinformation and the devisiveness of social media. I think everyone's used to very cheap reward functions pretty much across the board. So i guess what you're kind of pointing to with these reward functions is potentially more expensive to valuate them. Ye, there's also a question of agrede, whether the companies will want to do this. But i would hope at least that it would be a lot easier for the regulators to be like, yes, every one, you must train your d recommender systems to be optimizing for what humans would predict is good.
Transcript
Play full episode
