AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

OVLR: Efficient, Scalable, and Robust Training via Output-Level Variance-Reduced Likelihood Ratio

Minhao Zou, Tao Ren, Jinyang Jiang, Rui Tao, Zehao Li, Jiale Fu, Hui Shao, Xianhua Liu, Yijie Peng

ICML 2026 · Probabilistic and scientific modeling › Bayesian optimization · approach: Offline black-box optimization with surrogates

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 antithetic sampling (ablation)OVLR, ResNet-18 · CIFAR-100top-1 accuracy (%): 36.62 -> 60.06Improved
add antithetic sampling (ablation)OVLR, ResNet-18 · TinyImageNettop-1 accuracy (%): 18.78 -> 46.90Improved
add antithetic sampling (ablation)OVLR, ResNet-18 · CIFAR-10top-1 accuracy (%): 88.15 -> 88.47Improved
increase output-level repetition from n=1 to n=500 (ablation)OVLR, ResNet-18 · CIFAR-100top-1 accuracy (%): 1.85 -> 62.45Improved
vary noise scale sigma from 0.1 to 5 (ablation)OVLR with AS and n>=200 · CIFAR-100top-1 accuracy (%): 59.32 -> 58.98No clear effect
use Rademacher noise instead of Gaussian (ablation)OVLR, ResNet-18 · CIFAR-10best accuracy (%): 87.62 -> 26.95Hurt
move perturbation from parameters to outputs (vanilla LR to OVLR) (ablation)likelihood-ratio training · MNIST SimpleCNNaccuracy (%): 82.70 -> 96.80Improved
move perturbation from input level to output level (GLR-style to OVLR) (ablation)likelihood-ratio training · MNIST SimpleCNNaccuracy (%): 94.60 -> 96.80Improved
replace BP with OVLRViT-B/16 · CIFAR-100accuracy (%): 86.79 -> 88.17Improved
replace BP with OVLRResNet-18 · CIFAR-100accuracy (%): 63.10 -> 62.10Hurt
optimize hard 0-1 loss with OVLR instead of CE under 60% label noiseResNet-18 · CIFAR-10peak accuracy (%): 66.3 -> 73.0Improved
add pure BP gradient mixing (alpha=0.5) to OVLR (ablation)hybrid estimator · MNISTaccuracy (%): 96.09 -> 96.20No 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