AI Paper Atlas

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

Quantum Robust Inner Minimization for Reinforcement Learning with Quadratic Speed-Up in Query Complexity

Hyun Kyu Lee, Joongheon Kim, Sung Whan Yoon

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Tree search and policy optimization

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 exhaustive scan with quantum minimum finding (QRIM) in inner minimizationClassical RRL · CartPole Swingup extrapolationmean discounted return: 39.86 -> 58.06Improved
replace exhaustive scan with QRIMClassical RRL · FrozenLake-8x8 extrapolationmean discounted return: −0.45 -> −0.37Improved
replace exhaustive scan with QRIMClassical RRL · FrozenLake-8x8 trainmean discounted return: 0.84 -> 0.98Improved
replace exhaustive scan with QRIMClassical RRL · CartPole Swingup trainmean discounted return: 61.03 -> 76.41Improved
replace exhaustive scan with QRIMClassical RRL · FrozenLake-8x8cumulative oracle calls: 11.45M -> 4.11MImproved
replace exhaustive scan with QRIMClassical RRL · CartPole Swingupcumulative oracle calls: 27.51M -> 5.65MImproved
add robust training over uncertainty setClassic RL · FrozenLake-8x8 extrapolationmean discounted return: −0.79 -> −0.37Improved
replace BAI-based inner search with QRIMBAI-RRL · FrozenLake-8x8cumulative oracle calls: 9.67M -> 4.11MImproved
run QRIM on real quantum hardware instead of noise-free simulationQRIM · FrozenLake-8x8 extrapolationmean discounted return: −0.37 -> −0.52Hurt
increase QAE tolerance from 0.01 to 0.20 (ablation)QRIM · CartPole Swingup trainmean evaluation return: 76.41 -> 56.14Hurt
increase QAE tolerance from 0.01 to 0.20 (ablation)QRIM · CartPole Swingup extrapolationmean evaluation return: 58.06 -> 27.04Hurt
increase QAE tolerance from 0.01 to 0.10 (ablation)QRIM · CartPole Swingup trainmean evaluation return: 76.41 -> 72.96No clear effect

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