NeurIPS Political Economy of Reinforcement Learning Panel

Abstract

Panel discussion on the near-term policy concerns related to reinforcement learning (RL). What happens if RL algorithms trained on objectives related to human behavior attempt to manipulate human behavior to satisfy those objectives? Other panelists include Michael Dennis, Mireille Hildebrandt, and Salome Viljoen.

Date
2021
Location
Virtual
Natasha Jaques
Natasha Jaques

My research is focused on Social Reinforcement Learning–developing algorithms that use insights from social learning to improve AI agents’ learning, generalization, coordination, and human-AI interaction.