AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

End-to-End Urban Driving by Imitating a Reinforcement Learning Coach

Zhejun Zhang, Alexander Liniger, Dengxin Dai, Fisher Yu, Luc Van Gool

ICCV 2021 · Reinforcement learning › Visual and representation RL · approach: RL for combinatorial routing · 239 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
replace Autopilot with Roach as the expert (ablation)CILRS+DAGGER with L1 action loss · NoCrash-busy new town & new weatherdriving score (%): 43 -> 66Improved
add action-distribution KL loss instead of L1 action loss (ablation)CILRS+DAGGER supervised by Roach · NoCrash-busy new town & new weatherdriving score (%): 66 -> 79Improved
add latent feature matching loss with command-augmented input (ablation)CILRS+DAGGER with KL loss · NoCrash-busy new town & new weatherdriving score (%): 79 -> 88Improved
add latent feature matching loss with command-augmented input (ablation)CILRS+DAGGER with KL loss · NoCrash-busy new town & new weathersuccess rate (%): 74 -> 87Improved
add feature matching without command input on NoCrash (ablation)CILRS with KL loss · NoCrashdriving score (%)Hurt
add feature matching on LeaderBoard (ablation)CILRS with KL loss · offline LeaderBoarddriving score (%)Improved
add KL action distribution loss instead of L1 action loss (ablation)CILRS supervised by Roach · offline LeaderBoarddriving score (%)No clear effect
add value supervision on top of feature matching (ablation)CILRS with KL and feature loss · NoCrash-busydriving score (%)Improved
add value supervision without feature matching (ablation)CILRS with KL loss · NoCrash-busydriving score (%)No clear effect
add Beta distribution to PPO (ablation)Roach (PPO+exp) · NoCrash-dense new town & new weathersuccess rate (%): 77 -> 87Improved
add exploration loss to PPO with Beta (ablation)Roach (PPO+beta) · NoCrash-dense new town & new weathersuccess rate (%): 87 -> 83No clear effect
add exploration loss to PPO with Beta (ablation)PPO+beta · CARLA training stabilityepisode returnImproved

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