Research map / Architectures and efficiency / RNNs and normalization
The Diversified Ensemble Neural Network
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add diversified ensemble layer (N=4) | VGG16 · CIFAR-100 | top-1 accuracy: 70.48% -> 73.31% | Helped |
| add diversified ensemble layer vs random ensemble | VGG16 · CIFAR-100 | top-1 accuracy: 71.02% -> 73.31% | Helped |
| add diversified ensemble layer | NN · KDD10 | accuracy: 77.14% -> 79.19% | Helped |
| add diversified ensemble layer | NN · Higgs Boson | accuracy: 82.83% -> 83.08% | Mixed |
| add diversified ensemble layer | NN · Credit | MAE: 14.9901 -> 14.1028 | Helped |
| add diversified ensemble layer instead of random ensemble | ResNet-101 · CIFAR-10 | training time: 45m vs 15m | Helped |
| add diversified ensemble layer | BiLSTM · 20NG | accuracy: 73.18 -> 77.30 | Helped |
| add diversified ensemble layer | Text-GCN · MR | accuracy: 77.14 -> 78.37 | Helped |
| increase number of modules N from 1 to 4 (ablation) | DEns · CIFAR-10 | top-1 accuracy: 93.56% -> 94.89% | Helped |
| increase N from 4 to 6 (ablation) | DEns · CIFAR-10 | top-1 accuracy: 94.89% -> 94.66% | Hurt |
| remove diversity loss (alpha=0 vs adaptive alpha) (ablation) | E-Text-GCN · 20NG | accuracy: 87.06 vs 86.01 | Hurt |
| large fixed alpha=1 vs adaptive alpha (ablation) | E-Text-GCN · MR | accuracy: 78.37 vs 70.75 | Hurt |
Closest papers by meaning
- Prune and Tune Ensembles: Low-Cost Ensemble Learning with Sparse Independent Subnetworks (AAAI 2022 · similarity 0.89)
- Improving Ensemble Distillation With Weight Averaging and Diversifying Perturbation (ICML 2022 · similarity 0.89)
- Ensemble Distribution Distillation via Flow Matching (ICML 2025 · similarity 0.86)
- Theoretical Limitations of Ensembles in the Age of Overparameterization (ICML 2025 · similarity 0.85)
- United We Stand: Using Epoch-Wise Agreement of Ensembles to Combat Overfit (AAAI 2024 · similarity 0.84)
- Ensemble Pruning for Out-of-distribution Generalization (ICML 2024 · similarity 0.84)
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.