Avoiding Tampering Incentives in Deep RL via Decoupled Approval

17 November 2020

Papers citing "Avoiding Tampering Incentives in Deep RL via Decoupled Approval"

5 / 5 papers shown

Title
Sailing AI by the Stars: A Survey of Learning from Rewards in Post-Training and Test-Time Scaling of Large Language Models Xiaobao Wu LRM 72 1 0 05 May 2025
MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking Sebastian Farquhar Vikrant Varma David Lindner David Elson Caleb Biddulph Ian Goodfellow Rohin Shah 82 1 0 22 Jan 2025
Defining and Characterizing Reward Hacking Joar Skalse Nikolaus H. R. Howe Dmitrii Krasheninnikov David M. Krueger 57 54 0 27 Sep 2022
AI safety via debate G. Irving Paul Christiano Dario Amodei 199 199 0 02 May 2018
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks Chelsea Finn Pieter Abbeel Sergey Levine OOD 278 11,677 0 09 Mar 2017