AI Paper Atlas

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

Learning to Optimize Job Shop Scheduling Under Structural Uncertainty

Rui Zhang, Jianwei Niu, Xuefeng Liu, Shaojie Tang, Jing Yuan

AAAI 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Off-policy actor-critic with expressive policies

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
train critic on hindsight-reconstructed deterministic state (asymmetric actor-critic) (ablation)UP-AAC with UPM · JSSP 10x10, 1 branchaverage makespan: 934.49 -> 846.26Improved
train critic on hindsight-reconstructed deterministic state (ablation)UP-AAC with UPM · JSSP 10x20, 3 branchesaverage makespan: 1434.63 -> 1334.83Improved
add Uncertainty Perception Model (ablation)AAC · JSSP 10x20, 3 branchesaverage makespan: 1369.99 -> 1334.83 (CVaR 1508.02 -> 1457.83)Improved
add UPM to a standard (symmetric) actor-critic without AAC (ablation)standard AC · JSSP 10x10, 1 branchaverage makespan: 930.98 -> 934.49Hurt
replace standard actor-critic with UP-AACstandard AC · JSSP large instances 20x20, 3average makespan: 1922.53 -> 1884.30Improved
UP-AAC vs best heuristic baselines on large instancesMWKR/MOR · JSSP large instancesaverage makespan: 1704.50 -> 1670.13 (15x20, 3)Improved

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