Research map / Label-efficient and robust learning / Recognition and long-tail
Understanding Deep Representation Learning via Layerwise Feature Compression and Discrimination
Peng Wang, Xiao Li, Can Yaras, Zhihui Zhu, Laura Balzano, Wei Hu, Qing Qu
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 more projection head layers (1 to 5) during pre-training (ablation) | ResNet18 pre-trained on CIFAR-100 · CIFAR-10 linear probe transfer | transfer accuracy | Improved |
| add depth by stacking extra linear or MLP layers on a 2-layer MLP (3 to 9 layers) (ablation) | 2-layer MLP · FashionMNIST and CIFAR-10 | test accuracy | Improved |
| replace deep nonlinear MLP layers with a 5-layer linear network after a 3-layer MLP (ablation) | 8-layer MLP · CIFAR-10 | within-class compression and between-class discrimination | No clear effect |
Closest papers by meaning
- Gradients as Features for Deep Representation Learning (ICLR 2020 · similarity 0.85)
- Unsupervised Binary Representation Learning with Deep Variational Networks (IJCV 2019 · similarity 0.84)
- Towards Image Understanding from Deep Compression Without Decoding (ICLR 2018 · similarity 0.84)
- Compressing Neural Networks: Towards Determining the Optimal Layer-wise Decomposition (NeurIPS 2021 · similarity 0.83)
- On Compressing Deep Models by Low Rank and Sparse Decomposition (CVPR 2017 · similarity 0.83)
- Neural Representations Reveal Distinct Modes of Class Fitting in Residual Convolutional Networks (AAAI 2023 · 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.
