Research map / Label-efficient and robust learning / Recognition and long-tail
Reducing Data Complexity Using Autoencoders With Class-Informed Loss Functions
David Charte, Francisco Charte, Francisco Herrera
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add class-informed Slicer penalty to the autoencoder (ablation) | AE · MLP classifier F-score | average rank (lower is better): 4.731 -> 2.000 | Improved |
| add class-informed Slicer penalty to the autoencoder (ablation) | AE · kNN classifier F-score | average rank (lower is better): 4.115 -> 2.212 | Improved |
| add class-informed Slicer penalty to the autoencoder (ablation) | AE · SVM classifier F-score | average rank (lower is better): 3.538 -> 2.692 | Improved |
| add KLD-based Skaler penalty to the autoencoder (ablation) | AE · MLP classifier F-score | average rank (lower is better): 4.731 -> 2.596 | Improved |
| add Fisher-ratio Scorer penalty to the autoencoder (ablation) | AE · MLP classifier F-score | average rank (lower is better): 4.731 -> 3.865 | Improved |
| add Fisher-ratio Scorer penalty to the autoencoder (ablation) | AE · SVM classifier F-score | average rank (lower is better): 3.538 -> 2.923 | Improved |
| add KLD-based Skaler penalty to the autoencoder (ablation) | AE · complexity metric F1 | average rank (lower is better): 4.038 -> 1.577 | Improved |
| add Slicer penalty to the autoencoder (ablation) | AE · complexity metric F3 | average rank (lower is better): 5.654 -> 2.231 | Improved |
| use Slicer instead of PCA (ablation) | PCA · kNN classifier F-score | average rank (lower is better): 3.096 -> 2.212 | Improved |
| use Slicer instead of PCA (ablation) | PCA · complexity metric L2 | average rank (lower is better): 2.712 -> 2.846 | Hurt |
| use Skaler (F1-separability winner) instead of Slicer (ablation) | Slicer · complexity metric F1 vs classifier | average rank (lower is better): 1.577 vs 2.346 on F1; 2.596 vs 2.000 on MLP F-score | Mixed |
| use Slicer instead of Isomap | Isomap · kNN classifier F-score | average rank (lower is better): 4.904 -> 2.212 | Improved |
Closest papers by meaning
- Theoretical Insights Into Multiclass Classification: A High-dimensional Asymptotic View (NeurIPS 2020 · similarity 0.83)
- Escaping Saddle Points for Effective Generalization on Class-Imbalanced Data (NeurIPS 2022 · similarity 0.82)
- Restoring balance: principled under/oversampling of data for optimal classification (ICML 2024 · similarity 0.82)
- Selective Deep Autoencoder for Unsupervised Feature Selection (AAAI 2024 · similarity 0.81)
- Balancing the Scales: A Theoretical and Algorithmic Framework for Learning from Imbalanced Data (ICML 2025 · similarity 0.81)
- Breaking Class Barriers: Efficient Dataset Distillation via Inter-Class Feature Compensator (ICLR 2025 · similarity 0.81)
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.
