Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:1912.01683v2

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Artificial Intelligence

arXiv:1912.01683v2 (cs)
[Submitted on 3 Dec 2019 (v1), revised 19 Jan 2020 (this version, v2), latest version 28 Jan 2023 (v10)]

Title:Optimal Farsighted Agents Tend to Seek Power

Authors:Alexander Matt Turner
View a PDF of the paper titled Optimal Farsighted Agents Tend to Seek Power, by Alexander Matt Turner
View PDF
Abstract:Some researchers have speculated that capable reinforcement learning (RL) agents pursuing misspecified objectives are often incentivized to seek resources and power in pursuit of those objectives. An agent seeking power is incentivized to behave in undesirable ways, including rationally preventing deactivation and correction. Others have voiced skepticism: humans seem idiosyncratic in their urges to power, which need not be present in the agents we design. We formalize a notion of power within the context of finite deterministic Markov decision processes (MDPs). We prove that, with respect to a neutral class of reward function distributions, optimal policies tend to seek power over the environment.
Comments: 11 pages, 24 pages with appendix
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:1912.01683 [cs.AI]
  (or arXiv:1912.01683v2 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.1912.01683
arXiv-issued DOI via DataCite

Submission history

From: Alexander Turner [view email]
[v1] Tue, 3 Dec 2019 20:45:49 UTC (845 KB)
[v2] Sun, 19 Jan 2020 19:25:51 UTC (43 KB)
[v3] Mon, 13 Apr 2020 14:56:27 UTC (47 KB)
[v4] Tue, 14 Apr 2020 22:13:56 UTC (47 KB)
[v5] Fri, 5 Jun 2020 22:41:45 UTC (40 KB)
[v6] Wed, 2 Dec 2020 21:40:39 UTC (54 KB)
[v7] Tue, 1 Jun 2021 16:59:04 UTC (533 KB)
[v8] Sat, 23 Oct 2021 20:12:14 UTC (470 KB)
[v9] Fri, 3 Dec 2021 17:27:16 UTC (469 KB)
[v10] Sat, 28 Jan 2023 19:15:05 UTC (478 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Optimal Farsighted Agents Tend to Seek Power, by Alexander Matt Turner
  • View PDF
  • Other Formats
view license
Current browse context:
cs.AI
< prev   |   next >
new | recent | 2019-12
Change to browse by:
cs

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar

3 blog links

(what is this?)

DBLP - CS Bibliography

listing | bibtex
Alexander Matt Turner
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack