AI Paper Atlas

Research map / Multi-agent systems and games / Games and equilibria

Consequences of Misaligned AI

Simon Zhuang, Dylan Hadfield-Menell

NeurIPS 2020 · Multi-agent systems and games › Games and equilibria · approach: Learning with strategic agents · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Optimize a fixed proxy that omits at least one attributePrincipal-agent model · Resource-constrained state space satisfying the compactness conditionprincipal utilityHurt
Keep unmentioned attributes constant during proxy optimizationImpact-minimizing robot · Principal-agent modelprincipal utilityImproved
Interact regularly, using an efficient robot and a proxy over the most sensitive attributesInteractive robot · Within a neighborhood of the starting stateprincipal utilityImproved
Immediately send the Off signal under worst-case robot behaviorInteractive principal-agent model · Worst-case optimizationprincipal utility: 0 utilityNo clear effect
Combine regular interaction with keeping unmentioned attributes constant, and select the most and least sensitive attributesInteractive impact-minimizing robot · Principal-agent modelprincipal utilityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy