Many recommender systems are based on optimizing a linear weighting of
d...
As social media continues to have a significant influence on public opin...
Online platforms regularly conduct randomized experiments to understand ...
Most recommendation engines today are based on predicting user engagemen...
It is often difficult to hand-specify what the correct reward function i...
As machine learning models are increasingly used for high-stakes decisio...
Consequential decision-making incentivizes individuals to adapt their
be...
It is incredibly easy for a system designer to misspecify the objective ...
Consequential decision-making typically incentivizes individuals to beha...
We show through theory and experiment that gradient-based explanations o...
Autonomous agents optimize the reward function we give them. What they d...
Teachers intentionally pick the most informative examples to show their
...
Intuitively, obedience -- following the order that a human gives -- seem...