AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

Resolving Copycat Problems in Visual Imitation Learning via Residual Action Prediction

Chia-Chi Chuang, Donglin Yang, Chuan Wen, Yang Gao

ECCV 2022 · Reinforcement learning › Imitation learning · approach: Behavior cloning theory and models · 12 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
add action residual prediction memory module with policy streamBCOH (CILRS) · CARLA NoCrash Dense (train town & weather)#SUCCESS: 34.1 -> 52.0Improved
add stop-gradient on memory feature (ablation)OURS · CARLA NoCrash Dense (train town & weather)#SUCCESS: 45.0 -> 52.0Improved
train memory module on residual a_t - a_{t-1} instead of a_t (ablation)OURS · CARLA NoCrash Dense (train town & weather)#SUCCESS: 41.3 -> 52.0Improved
train memory module on residual instead of a_{t-1} (ablation)OURS · CARLA NoCrash Dense (train town & weather)#SUCCESS: 47.0 -> 52.0Improved
add policy stream with current observation to memory module (ablation)OURS · CARLA NoCrash Dense (train town & weather)#SUCCESS: 0.0 -> 52.0Improved
add residual prediction branches a_t - a_{t-i} (2 branches) (ablation)OURS · CARLA NoCrash Dense (train town & weather)#SUCCESS: 52.0 -> 52.6No clear effect
use 4 residual branches (ablation)OURS · CARLA NoCrash Dense (train town & weather)#SUCCESS: 52.0 -> 50.7Hurt
use two-stream architecture without residual objective (ablation)BCOH · CARLA NoCrash Dense (train town & weather)#SUCCESS: 34.1 -> 23.7Hurt
use two-stream architecture without residual objective (ablation)Keyframe · CARLA NoCrash Dense (train town & weather)#SUCCESS: 41.9 -> 38.7Hurt
use residual memory method instead of BCOH on intervention testBCOH · CARLA validation setbehavior change %: 40.88 -> 16.18Improved
use residual memory feature, measured by previous-action regression MSEBCOH · CARLA validation setMSE loss x10^-2: 4.83 -> 7.50Improved
use residual memory method instead of BCOH on green light passingBCOH · CARLA NoCrash Densegreen light passing %: 30.6 -> 66.3Improved

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