AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

EDGE: Explaining Deep Reinforcement Learning Policies

Wenbo Guo, Xian Wu, Usmann Khan, Xinyu Xing

NeurIPS 2021 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 18 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 EDGE explanations instead of best baseline to pick attack stepsexplanation-guided DRL agent intervention · Pong attackwin rate change (points): -30.33 -> -65.47Improved
use EDGE explanations instead of Rudder to pick attack stepsexplanation-guided DRL agent intervention · You-Shall-Not-Pass attackwin rate change (points): -32.53 -> -35.13No clear effect
use EDGE explanations instead of Attention RNN for attackexplanation-guided DRL agent intervention · Kick-And-Defend attackwin rate change (points): -41.20 -> -43.47No clear effect
use EDGE explanations for policy patching vs Rationale Netexplanation-guided DRL agent intervention · Pong patchingwin rate change (points): -3.66 -> +2.75Improved
use EDGE explanations for policy patching vs Rudderexplanation-guided DRL agent intervention · You-Shall-Not-Pass patchingwin rate change (points): +1.76 -> +2.91Improved
use EDGE explanations for policy patching vs Saliencyexplanation-guided DRL agent intervention · Kick-And-Defend patchingwin rate change (points): +1.17 -> +1.21No clear effect
blind victim at EDGE-identified critical steps vs Rudderexplanation-guided DRL agent intervention · You-Shall-Not-Pass robustifyingvictim win rate change (points): +8.54 -> +35.30Improved
EDGE explanation vs Attention/Rationale Net/saliency/Rudderexplanation-guided DRL agent intervention · Pong, MuJoCo-Y, MuJoCo-Kexplanation fidelity scoreImproved
self-explainable Attention and Rationale Net vs post-training explanationsexplanation-guided DRL agent intervention · Pong, MuJoCoexplanation fidelity scoreNo clear effect

Later papers that compared with this method

2 later accepted papers report EDGE as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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