AI Paper Atlas

Research map / Optimization / Optimizers

Towards Dynamic Interleaving Optimizers

Yile Chen, Zeyi Wen, Jian Chen, Jin Huang

ICLR 2026 · Optimization › Optimizers · approach: Second-order and structured preconditioning

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
switch optimizers dynamically with DOIT instead of a static optimizerViT full training · cifar10accuracy (%): 97.58 -> 98.04Improved
use DOIT on a complex imbalanced datasetfull training · imagenet-atop-1 accuracy (%): 16.31 -> 18.47Improved
use DOIT instead of hybrid optimizers (SWATS, Padam, AdaBound)RoBERTa LoRA · qqpaccuracy (%): 73.04 -> 85.57 (vs Padam)Improved
use DOIT on additional taskstranslation, regression, detection · wmt14 / eunite / cocoBLEU / R2 / AP50: 28.24 -> 28.68 BLEU; 71.81 -> 73.12 R2; 57.44 -> 59.21 AP50Improved
use DOIT instead of LR schedulers (stepLR, cosLR)full training · mnistaccuracy (%): 99.46 -> 99.51No clear effect
add DOIT scheduling overheadtraining time · ResNet18 cifar10time overheadNo clear effect
replace DOIT components (random/periodic switching, no transferability weight, uniform init, random projection/UMAP) (ablation)DOIT · ablation suiteaccuracyImproved
expand optimizer/hyperparameter search space (ablation)DOIT · uspsaccuracy and timeNo 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