Combating Deep Reinforcement Learning ’ s Sisyphean Curse with Reinforcement Learning

By Zachary Chase Lipton, Abhishek Kumar, Jianfeng Gao, Lihong Li, Li Deng zlipton. 2017

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Owain Evans
2. Paul Trichelair
3. Romain Laroche
4. William Saunders
5. Andreas Stuhlmüller
6. Girish Sastry
7. Tom Everitt
8. Jan Leike

Citations in the corpus

• Safe Policy Improvement with Baseline Bootstrapping (2017, background): “, 2016) or in the catastrophe avoidance scenarios (Geibel & Wysotzki, 2005; Lipton et al., 2016).”
• Trial without Error: Towards Safe Reinforcement Learning via Human Intervention (2017, background): “The Blocker never stops overseeing the agent, which prevents catastrophes even if the agent exhibits random exploration or catastrophic forgetting [6].”