AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

How to Lose Inherent Counterfactuality in Reinforcement Learning

Ezgi Korkmaz

ICLR 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep value learning

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 epsilon-invariance (certified robust) trainingDDQN policy · BankHeistAUC performance drop under second-best action modification: 0.191 -> 0.449Hurt
add epsilon-invariance trainingDDQN policy · RoadRunnerAUC performance drop under second-best action modification: 0.247 -> 0.414Hurt
add epsilon-invariance trainingDDQN policy · FreewayAUC performance drop under second-best action modification: 0.302 -> 0.351Hurt
add epsilon-invariance trainingDDQN policy · BankHeist, RoadRunner, FreewayAUC performance drop under worst-action modification: BankHeist 0.398 -> 0.311Mixed
add epsilon-invariance trainingDDQN policy · ALE gamesQ-value of best actionHurt

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