Research map / Optimization / Implicit bias and generalization
Regularization in ResNet with Stochastic Depth
Soufiane Hayou, Fadhel Ayed
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 |
|---|---|---|---|
| SenseMode instead of Uniform mode at budget 0.1 | Stable ResNet110 SD · CIFAR-10 | test error (%): 17.2 -> 15.4 | Improved |
| SenseMode instead of Uniform mode at budget 0.1 | Stable ResNet110 SD · CIFAR-100 | test error (%): 55.2 -> 51.3 | Improved |
| SenseMode instead of Uniform mode at budget 0.4 | Stable ResNet110 SD · CIFAR-100 | test error (%): 30.9 -> 28.5 | Improved |
| SenseMode instead of Uniform mode at budget 0.4 | Stable ResNet110 SD · CIFAR-10 | test error (%): 7.4 -> 7.3 | No clear effect |
| SenseMode instead of Uniform mode at budget 0.7 | Stable ResNet110 SD · CIFAR-10 | test error (%): 5.9 -> 7.3 | Hurt |
| SenseMode instead of Uniform mode at budget 0.8 | Stable ResNet110 SD · CIFAR-100 | test error (%): 25.5 -> 30.0 | Hurt |
| Linear mode instead of Uniform mode at budget 0.5 | Stable ResNet110 SD · CIFAR-10 | test error (%): 6.8 -> 9.1 | Hurt |
| Linear mode instead of Uniform mode at budget 0.5 | Stable ResNet110 SD · CIFAR-100 | test error (%): 28.4 -> 36.5 | Hurt |
| Linear mode instead of Uniform mode at budget 0.9 | Stable ResNet110 SD · CIFAR-100 | test error (%): 25.5 -> 25.5 | No clear effect |
| Uniform stochastic depth at budget 0.9 versus full network | Stable ResNet110 · CIFAR-10 | test error (%): 6.37 -> 5.7 | Improved |
| Uniform stochastic depth at budget 0.9 versus full network | Stable ResNet110 · CIFAR-100 | test error (%): 26.5 -> 25.5 | Improved |
| reduce training budget from 0.9 to 0.1 with Uniform mode | Stable ResNet110 SD · CIFAR-10 | test error (%): 5.7 -> 17.2 | Hurt |
Closest papers by meaning
- Deep Networks with Stochastic Depth (ECCV 2016 · similarity 0.88)
- Neural Collapse is Globally Optimal in Deep Regularized ResNets and Transformers (NeurIPS 2025 · similarity 0.87)
- Are ResNets Provably Better than Linear Predictors? (NeurIPS 2018 · similarity 0.87)
- Overparameterization of Deep ResNet: Zero Loss and Mean-field Analysis (JMLR 2022 · similarity 0.87)
- Regularizing Deep Neural Networks by Noise: Its Interpretation and Optimization (NeurIPS 2017 · similarity 0.87)
- A Mean Field Analysis Of Deep ResNet And Beyond: Towards Provably Optimization Via Overparameterization From Depth (ICML 2020 · similarity 0.87)
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.
