research
∙
04/13/2023
Power-seeking can be probable and predictive for trained agents
Power-seeking behavior is a key source of risk from advanced AI, but our...
share
research
∙
10/04/2022
Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals
The field of AI alignment is concerned with AI systems that pursue unint...
research
∙
11/17/2020
Avoiding Tampering Incentives in Deep RL via Decoupled Approval
How can we design agents that pursue a given objective when all feedback...
research
∙
11/17/2020
REALab: An Embedded Perspective on Tampering
This paper describes REALab, a platform for embedded agency research in ...
research
∙
10/15/2020
Avoiding Side Effects By Considering Future Tasks
Designing reward functions is difficult: the designer has to specify wha...
research
∙
06/20/2019
Modeling AGI Safety Frameworks with Causal Influence Diagrams
Proposals for safe AGI systems are typically made at the level of framew...
research
∙
06/04/2018
Measuring and avoiding side effects using relative reachability
How can we design reinforcement learning agents that avoid causing unnec...
research
∙
11/27/2017
AI Safety Gridworlds
We present a suite of reinforcement learning environments illustrating v...
research
∙
05/23/2017