atlas

Research map / Architectures and efficiency / RNNs and normalization

Towards Inference Efficient Deep Ensemble Learning

AAAI 2023 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 9 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Use learned sequential halting and joint model-selector optimization instead of averaging all modelsResNet-18 ensemble · CIFAR-10Top-1 accuracy (%): +0.19%Helped
Use learned sequential halting rather than running the full average ensembleResNet-18 ensemble · CIFAR-10Cost: 56.0%Helped
Use IRENE instead of average ensembleResNet-32 ensemble · CIFAR-10Top-1 accuracy (%): 94.46 -> 94.46No clear effect
Use IRENE instead of average ensembleResNet-18 ensemble · CIFAR-100Top-1 accuracy (%): 80.28 -> 80.10No clear effect
Optimize the selector with ensemble-performance, cost, and ranking objectives (ablation)IRENE · CIFAR-10UtilityHelped
Optimize base models jointly with the selector using the ranking objective (ablation)IRENE · CIFAR-10UtilityHelped
Add the ranking objective to base-model and selector training (ablation)IRENE · CIFAR-10Base-model lossHelped

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