Research map / Label-efficient and robust learning / Recognition and long-tail
AE2-Nets: Autoencoder in Autoencoder Networks
Check 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 |
|---|---|---|---|
| Use nested view-specific and multi-view autoencoders instead of the strongest compared baseline | AE2-Nets · handwritten classification, G80%/P20% through G20%/P80% | accuracy: 95.78 ± 0.46 -> 96.93 ± 0.71; 95.10 ± 0.64 -> 96.55 ± 0.66; 94.79 ± 0.58 -> 95.88 ± 0.71; 92.79 ± 0.51 -> 93.38 ± 0.49 | Helped |
| Use nested view-specific and multi-view autoencoders instead of the strongest compared baseline | AE2-Nets · Caltech101 classification, G80%/P20% through G20%/P80% | accuracy: 92.12 ± 0.58 -> 93.77 ± 1.35; 91.46 ± 0.70 -> 92.98 ± 1.37; 91.30 ± 0.48 -> 92.49 ± 0.72; 90.73 ± 0.38 -> 91.36 ± 0.69 | Helped |
| Use nested view-specific and multi-view autoencoders instead of the strongest compared baseline | AE2-Nets · ORL classification, G80%/P20% through G20%/P80% | accuracy: 93.13 ± 1.21 -> 97.88 ± 1.19; 91.75 ± 1.64 -> 96.00 ± 2.18; 85.45 ± 1.85 -> 92.20 ± 1.18; 57.38 ± 2.08 -> 70.16 ± 2.54 | Helped |
| Use nested view-specific and multi-view autoencoders instead of the strongest compared baseline | AE2-Nets · COIL20 classification, G80%/P20% through G20%/P80% | accuracy: 95.25 ± 1.06 -> 96.11 ± 1.10; 94.76 ± 0.77 -> 95.55 ± 0.87; 92.07 ± 0.61 -> 93.25 ± 0.73; 84.11 ± 1.10 -> 88.85 ± 0.72 | Helped |
| Use nested view-specific and multi-view autoencoders instead of the strongest compared baseline | AE2-Nets · CUB classification, G80%/P20% through G20%/P80% | accuracy: 83.08 ± 3.43 -> 85.83 ± 2.94; 82.44 ± 3.08 -> 84.00 ± 1.41; 81.53 ± 1.67 -> 82.67 ± 1.41; 78.58 ± 1.65 -> 80.17 ± 1.83 | Helped |
| Replace the nonlinear nested networks with linear projections | AE2-Nets · multi-view representation learning | reconstruction objective | Mixed |
| Increase the tradeoff parameter λ across the tested range | AE2-Nets · handwritten clustering | NMI | No clear effect |
Closest papers by meaning
- AET vs. AED: Unsupervised Representation Learning by Auto-Encoding Transformations Rather Than Data (CVPR 2019 · similarity 0.87)
- Supervised autoencoders: Improving generalization performance with unsupervised regularizers (NeurIPS 2018 · similarity 0.86)
- Deep Incomplete Multi-view Learning via Cyclic Permutation of VAEs (ICLR 2025 · similarity 0.84)
- Improved Baselines with Representation Autoencoders (NeurIPS 2026 · similarity 0.83)
- A Statistical Analysis of Wasserstein Autoencoders for Intrinsically Low-dimensional Data (ICLR 2024 · similarity 0.83)
- Exploring Target Representations for Masked Autoencoders (ICLR 2024 · similarity 0.83)
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.