Research map / Label-efficient and robust learning / Recognition and long-tail
Feature Selection Guided Auto-Encoder
Shuyang Wang, Zhengming Ding, Yun Fu
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 |
|---|---|---|---|
| train feature selection jointly with auto-encoder instead of applying it afterwards (AE+FS) (ablation) | FSAE (auto-encoder with feature selection) · COIL100 average | recognition rate (%): 88.28 -> 91.27 | Improved |
| joint feature selection training instead of AE+FS (ablation) | FSAE (auto-encoder with feature selection) · COIL100 20 objects | recognition rate (%): 91.73 -> 94.12 | Improved |
| joint feature selection training instead of AE+FS (ablation) | FSAE (auto-encoder with feature selection) · Caltech101 30 train | recognition rate (%): 75.02 -> 77.14 | Improved |
| joint feature selection training instead of AE+FS (ablation) | FSAE (auto-encoder with feature selection) · Caltech101 10 train | recognition rate (%): 61.83 -> 64.90 | Improved |
| post-hoc feature selection (AE+FS) versus plain sparse AE (ablation) | FSAE (auto-encoder with feature selection) · Caltech101 10 train | recognition rate (%): 62.92 -> 61.83 | Hurt |
| post-hoc feature selection (AE+FS) versus plain SAE (ablation) | FSAE (auto-encoder with feature selection) · COIL100 average | recognition rate (%): 87.71 -> 88.28 | Improved |
| FSAE versus best non-AE baseline SRRS (ablation) | FSAE (auto-encoder with feature selection) · COIL100 100 objects | recognition rate (%): 85.12 -> 86.75 | Improved |
| vary selection ratio m/z; best around 0.3-0.5 of hidden units (ablation) | FSAE (auto-encoder with feature selection) · COIL100 | recognition rate (%) | No clear effect |
| stack deeper layers (layer 3 and 4) of FSAE (ablation) | FSAE (auto-encoder with feature selection) · COIL100 | recognition rate (%) | Hurt |
Closest papers by meaning
- Selective Deep Autoencoder for Unsupervised Feature Selection (AAAI 2024 · similarity 0.87)
- SAND: One-Shot Feature Selection with Additive Noise Distortion (ICML 2025 · similarity 0.85)
- Indirectly Parameterized Concrete Autoencoders (ICML 2024 · similarity 0.83)
- Estimating Conditional Mutual Information for Dynamic Feature Selection (ICLR 2024 · similarity 0.82)
- PEAKS: Selecting Key Training Examples Incrementally via Prediction Error Anchored by Kernel Similarity (ICML 2025 · similarity 0.82)
- Contextual Feature Selection with Conditional Stochastic Gates (ICML 2024 · 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.
