paper-with-me

Papers

Understanding Agent Incentives using Causal Influence Diagrams. Part I: Single Action Settings

2019-02-26 · Tom Everitt, Pedro A. Ortega, Elizabeth Barnes, Shane Legg

Agents are systems that optimize an objective function in an environment. Together, the goal and the environment induce secondary objectives, incentives. Modeling the agent-environment interaction using causal influence diagrams, we can answer two fundamental questions about an agent's incentives directly from the graph: (1) which nodes can the agent have an incentivize to observe, and (2) which nodes can the agent have an incentivize to control? The answers tell us which information and influence points need extra protection. For example, we may want a classifier for job applications to not use the ethnicity of the candidate, and a reinforcement learning agent not to take direct control of its reward mechanism. Different algorithms and training paradigms can lead to different causal influence diagrams, so our method can be used to identify algorithms with problematic incentives and help in designing algorithms with better incentives.

📄 PDF Abstract BibTeX arXiv:1902.09980

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningReinforcement Learning (RL)

Similar Papers 제목 키워드 기반

Agent Incentives: A Causal Perspective

2021-02-02 · Tom Everitt, Ryan Carey, Eric Langlois, Pedro A Ortega 외

We present a framework for analysing agent incentives using causal influence diagrams. We establish that a well-known criterion for value of information is complete. We propose a new graphical criterion for value of cont…

Fairness

The Incentives that Shape Behaviour

2020-01-20 · Ryan Carey, Eric Langlois, Tom Everitt, Shane Legg

Which variables does an agent have an incentive to control with its decision, and which variables does it have an incentive to respond to? We formalise these incentives, and demonstrate unique graphical criteria for dete…

Fairness

Discovering Agents

2022-08-17 · Zachary Kenton, Ramana Kumar, Sebastian Farquhar, Jonathan Richens 외

Causal models of agents have been used to analyse the safety aspects of machine learning systems. But identifying agents is non-trivial -- often the causal model is just assumed by the modeler without much justification …

Causal Discovery

Path-Specific Objectives for Safer Agent Incentives

2022-04-21 · Sebastian Farquhar, Ryan Carey, Tom Everitt

We present a general framework for training safe agents whose naive incentives are unsafe. As an example, manipulative or deceptive behaviour can improve rewards but should be avoided. Most approaches fail here: agents m…

Modeling AGI Safety Frameworks with Causal Influence Diagrams

2019-06-20 · Tom Everitt, Ramana Kumar, Victoria Krakovna, Shane Legg

Proposals for safe AGI systems are typically made at the level of frameworks, specifying how the components of the proposed system should be trained and interact with each other. In this paper, we model and compare the m…