AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

AvE: Assistance via Empowerment

Yuqing Du, Stas Tiomkin, Emre Kiciman, Daniel Polani, Pieter Abbeel, Anca Dragan

NeurIPS 2020 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Human-AI coordination agents · 14 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
Use empowerment instead of goal inference when the goal set is large or misspecifiedGridworld assistance · Gridworld, human in centersuccess rate: 50% -> 100% (LG Set); 55% -> 100% (SG Set)Improved
Use empowerment rather than correctly specified goal inferenceGridworld assistance · Gridworld, human in center, SG Setmean steps: 2.57 -> 4.8Hurt
Replace exact empowerment with the empowerment-inspired proxy (ablation)Gridworld assistance · Gridworld, human in cornersuccess rate: 100% -> 99%Hurt
Replace exact empowerment with the empowerment-inspired proxy (ablation)Gridworld assistance · Gridworld, human in cornermean steps: 5.24 -> 8.79Hurt
Add empowerment bonus to the copilot reward (ablation)Lunar Lander copilot · Noisy pilotsuccessful landings: 8.5 -> 30.7Improved
Add empowerment bonus to the copilot reward (ablation)Lunar Lander copilot · Sensor pilotsuccessful landings: 2.8 -> 10.8Improved
Add empowerment bonus to the copilot reward (ablation)Lunar Lander copilot · Laggy pilotsuccessful landings: 38.4 -> 37.3No clear effect
Add empowerment-inspired bonus to the copilot reward (ablation)Human-copilot Lunar Lander team · Lunar Lander user studysuccess rate: $17 \pm 9$ -> 33 $\pm$ 7Improved
Add empowerment-inspired bonus to the copilot reward (ablation)Human-copilot Lunar Lander team · Lunar Lander user studyperceived task ease: 5.55 -> 6.4Improved
Increase the empowerment reward coefficient too much (ablation)Lunar Lander copilot · Lunar Lander user studysuccessful landingHurt

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