Research map / Optimization / Optimizers
Bayesian filtering unifies adaptive and non-adaptive neural network optimization methods
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use AdaBayes instead of Adam | ResNet-34 and DenseNet-121 · CIFAR-10 and CIFAR-100 | test error | Helped |
| Use AdaBayes instead of Adam | ResNet-34 and DenseNet-121 · CIFAR-10 and CIFAR-100 | test loss | Helped |
| Use AdaBayes instead of SGD | ResNet-34 and DenseNet-121 · CIFAR-10 | test error | Helped |
| Use AdaBayes instead of SGD | ResNet-34 and DenseNet-121 · CIFAR-10 and CIFAR-100 | test error | Mixed |
| Use AdaBayes-FP instead of AdaBayes (ablation) | ResNet-34 and DenseNet-121 · CIFAR-10 and CIFAR-100 | test error and loss | Helped |
| Use AdaBayes-FP instead of AdaBayes (ablation) | ResNet-34 and DenseNet-121 · CIFAR-10 classification | test error | No clear effect |
| Use AdaBayes or AdaBayes-FP instead of AdamW, AdaBound, or AMSBound | ResNet-34 and DenseNet-121 · CIFAR-10 and CIFAR-100 | test performance | No clear effect |
| Use SGD instead of adaptive methods | ResNet-34 and DenseNet-121 · CIFAR-10 and CIFAR-100 | test loss | Mixed |
Closest papers by meaning
- Optimization and Bayes: A Trade-off for Overparameterized Neural Networks (NeurIPS 2023 · similarity 0.88)
- Implicit Maximum a Posteriori Filtering via Adaptive Optimization (ICLR 2024 · similarity 0.87)
- Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold (NeurIPS 2025 · similarity 0.87)
- A Study of Bayesian Neural Network Surrogates for Bayesian Optimization (ICLR 2024 · similarity 0.87)
- A Comprehensive Framework for Analyzing the Convergence of Adam: Bridging the Gap with SGD (ICML 2025 · similarity 0.87)
- MADA: Meta-Adaptive Optimizers Through Hyper-Gradient Descent (ICML 2024 · similarity 0.86)
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.