9.7k citations · 11.2k across the 18 of their papers we have counts for
Showing 2018Show all
2 papers · 1 filter
cs.LG2018★ 39 cited
Reward learning from human preferences and demonstrations in Atari
Borja Ibarz, Jan Leike, Tobias Pohlen +3
To solve complex real-world problems with reinforcement learning, we cannot rely on manually specified reward functions. Instead, we can have humans communicate an objective to the…
stat.ML2018★ 29 cited
AI safety via debate
Geoffrey Irving, Paul Christiano, Dario Amodei
To make AI systems broadly useful for challenging real-world tasks, we need them to learn complex human goals and preferences. One approach to specifying complex goals asks humans…