AI Paper Atlas

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

Reward-Directed Score-Based Diffusion Models via q-Learning

Xuefeng Gao, Jiale Zha, Xun Yu Zhou

JMLR 2025 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep value learning · 16 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
increase score-estimator sample size m from 30 to 120 (ablation)q-learning diffusion · Swiss roll (beta=0)KL divergence to data: 0.33 -> 0.17Improved
increase sample size m from 30 to 150 at fixed reward 0.8 (ablation)q-learning diffusion · Swiss roll (reward 0.8)KL divergence to data: 0.88 -> 0.38Improved
increase sample size m from 150 to 300 (ablation)q-learning diffusion · Swiss roll (beta=0)KL divergence to data: 0.18 -> 0.17No clear effect
increase sample size m from 30 to 300 (t=1) (ablation)ratio score estimator · Swiss roll, t=1MSE of score estimate: 5.348×10^-2 -> 8.682×10^-3Improved
increase sample size m from 30 to 300 (t=4) (ablation)ratio score estimator · Swiss roll, t=4MSE of score estimate: 1.112×10^-3 -> 3.209×10^-4Improved
increase sample size m from 30 to 300 (t=1) (ablation)ratio score estimator · Swiss roll, t=1variance of score estimate: 5.036×10^-2 -> 4.458×10^-3Improved
evaluate at larger t (t=4 instead of t=1)ratio score estimator · Swiss roll, m=30MSE of score estimate: 5.348×10^-2 -> 1.112×10^-3Improved

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