Research map / Architectures and efficiency / RNNs and normalization
Deep Neural Nets with Interpolating Function as Output Activation
Bao Wang, Xiyang Luo, Zhen Li, Wei Zhu, Zuoqiang Shi, Stanley Osher
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 the vanilla output activation with WNLL | ResNet50 · CIFAR100 test set | error rate: 25.09% -> 20.45% | Improved |
| Replace the vanilla output activation with WNLL | PreActResNet50 · CIFAR100 test set | error rate: 25.95% -> 21.51% | Improved |
| Replace the vanilla output activation with WNLL | VGG16 · CIFAR100 test set | error rate: 28.59% -> 25.72% | Improved |
| Replace the vanilla output activation with WNLL | ResNet110 · SVHN test set | error rate: 2.55% -> 2.26% | Improved |
| Replace the vanilla output activation with WNLL | PreActResNet34 · SVHN test set | error rate: 3.66% -> 3.32% | Improved |
| Use WNLL output activation instead of softmax | Deep neural nets trained with 1000 or 10000 CIFAR10 examples · CIFAR10 test set | generalization error: up to 10 percent | Improved |
| Use WNLL output activation instead of softmax | Deep neural nets trained with insufficient data · CIFAR10 test set | generalization error across network depth | Mixed |
Closest papers by meaning
- Efficient Activation Function Optimization through Surrogate Modeling (NeurIPS 2023 · similarity 0.84)
- Better NTK Conditioning: A Free Lunch from (ReLU) Nonlinear Activation in Wide Neural Networks (NeurIPS 2025 · similarity 0.83)
- A Representer Theorem for Deep Neural Networks (JMLR 2019 · similarity 0.83)
- On the Hardness of Training Deep Neural Networks Discretely (AAAI 2025 · similarity 0.83)
- A generalized neural tangent kernel for surrogate gradient learning (NeurIPS 2024 · similarity 0.82)
- Provable Tempered Overfitting of Minimal Nets and Typical Nets (NeurIPS 2024 · similarity 0.82)
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.
