AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Causal Influence Detection for Improving Efficiency in Reinforcement Learning

Maximilian Seitzer, Bernhard Schölkopf, Georg Martius

NeurIPS 2021 · Reinforcement learning › Visual and representation RL · approach: Hierarchical skills and priors · 9 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
estimate influence with CAI instead of Entropy scoreinfluence detection · Slideaverage precision: 0.47 -> 0.98Improved
estimate influence with CAI instead of Transformer attention weightsinfluence detection · FetchPickAndPlaceAUC: 0.46 -> 0.97Improved
estimate influence with CAI instead of simulator Contactsinfluence detection · FetchPickAndPlaceAUC: 0.79 -> 0.97Improved
add CAI reward bonus with appropriate scale to DDPG+HER (ablation)DDPG+HER · FetchPickAndPlaceepisodes to 60% success: reaches 60% success four-times fasterImproved
use too large CAI bonus scale (ablation)DDPG+HER with CAI bonus · FetchPickAndPlacesuccess rateHurt
add active causal action exploration on top of epsilon-greedy random exploration (ablation)DDPG+HER · FetchPickAndPlacesample efficiencyImproved
combine reward bonus with active exploration (ablation)CAI exploration · FetchPickAndPlacesample efficiencyImproved
use CAI exploration instead of VIMEexploration · FetchPickAndPlacesample efficiencyImproved
use CAI prioritization instead of PERDDPG+HER · Fetch manipulation tasks (60% success)learning speed: 1.5-2.5x fasterImproved
use CAI prioritization instead of energy-based EBP (privileged)DDPG+HER · Fetch manipulation taskslearning speedNo clear effect
combine two or three of prioritization, bonus and active exploration (ablation)DDPG+HER · FetchPickAndPlacesample efficiency: 4-10x increaseImproved
increase observation noise for influence detection (ablation)CAI · FetchPickAndPlaceaverage precisionHurt

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