PickVidsTraining AI Without Writing A Reward Function, with Reward Modelling
Why it was recommended
I find this video by Rob Miles still the best presentation of the ingenious original 2017(!) paper
Strong recommendationcubefox · Original
What it covers
Robert Miles explains reward modelling, in which an AI learns what humans want from their preference comparisons instead of a hand-written reward function. He walks through. It is for someone who already knows the basic terms and wants the next step.
Video details
- Channel
- Robert Miles AI Safety
- Length
- 17 min 52 sec
- Year
- 2019
More in this topic
- What Happened to Horses Is Happening to Us
- Why Would AI Want to do Bad Things? Instrumental Convergence
- Veritasium: What Everyone Gets Wrong About AI and Learning – Derek Muller Explains
- Intelligence and Stupidity: The Orthogonality Thesis
- We're building a dystopia just to make people click on ads | Zeynep Tufekci
- Intro to AI Safety, Remastered