Research map / Architectures and efficiency / CNN design
Augmenting Deep Classifiers with Polynomial Neural Networks
Grigorios G. Chrysos, Markos Georgopoulos, Jiankang Deng, Jean Kossaifi, Yannis Panagakis, Anima Anandkumar
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 |
|---|---|---|---|
| replace ResNet18 blocks with PDC keeping channels | ResNet18 · CIFAR100 | accuracy: 0.756 -> 0.773 | Improved |
| shrink PDC channels to compress the model | PDC · CIFAR100 | accuracy: 0.756 -> 0.760 | Improved |
| increase polynomial degree of PDC blocks from two to four (ablation) | PDC · CIFAR100 | accuracy: 0.765 -> 0.771 | Improved |
| add first-degree term and unshared factors to disentangled non-local block (ablation) | DNL · CIFAR100 (non-local blocks) | accuracy: 0.771 -> 0.773 | Improved |
| add fourth-degree squeeze-excitation-like term to PDC-NL[3] (ablation) | PDC-NL[3] · CIFAR100 (non-local blocks) | accuracy: 0.773 -> 0.779 | Improved |
| replace non-local block with PDC-NL[4] | ResNet18 · ImageNet (non-local blocks) | top-1 accuracy: 0.702 -> 0.716 | Improved |
| use PDC instead of ResNet18 or SENet on CIFAR10 | ResNet18 · CIFAR10 | accuracy: 0.945 -> 0.946 | No clear effect |
| use PDC instead of ResNet18 on speech classification | ResNet18 · Speech Commands | accuracy: 0.977 -> 0.978 | No clear effect |
| use PDC-param instead of ResNet18 under strong class imbalance | ResNet18 · CIFAR10-LT (IF 200) | accuracy: 0.645 -> 0.665 | Improved |
| use PDC-comp instead of ResNet18 with very limited data | ResNet18 · CIFAR10 (50 samples per class) | accuracy: 0.347 -> 0.426 | Improved |
| use PDC polynomial blocks instead of plain ResNet18 when activations are removed (ablation) | ResNet18 · CIFAR100 (no activation functions) | accuracy: 0.168 -> 0.689 | Improved |
| raise degree of Pi-net blocks from two to four without activations (ablation) | Pi-net-ResNet · CIFAR100 (no activation functions) | accuracy: 0.667 -> 0.626 | Hurt |
Closest papers by meaning
- Regularization of Polynomial Networks for Image Recognition (CVPR 2023 · similarity 0.88)
- Deep Polynomial Neural Networks (TPAMI 2021 · similarity 0.88)
- P-nets: Deep Polynomial Neural Networks (CVPR 2020 · similarity 0.87)
- Identifiability of Deep Polynomial Neural Networks (NeurIPS 2025 · similarity 0.86)
- On the Expressive Power of Deep Polynomial Neural Networks (NeurIPS 2019 · similarity 0.86)
- Multilinear Operator Networks (ICLR 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.
