Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning
Cuong Pham, Cuong C. Nguyen, Trung Le, Dinh Phung, Gustavo Carneiro, Thanh-Toan Do
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 parameter and feature diversity losses to Bayesian mutual learning (ablation) | Bayesian mutual learning · CIFAR-100 | top-1 accuracy: 68.44 -> 69.39 | Improved |
| Add parameter diversity loss, without feature diversity loss (ablation) | Bayesian mutual learning · CIFAR-100 | top-1 accuracy: 68.44 -> 69.00 | Improved |
| Add feature diversity loss, without parameter diversity loss (ablation) | Bayesian mutual learning · CIFAR-100 | top-1 accuracy: 68.44 -> 68.81 | Improved |
| Add parameter diversity loss to logits-only mutual learning (ablation) | Bayesian mutual learning, ResNet20 initialized from scratch · CIFAR-100 | top-1 accuracy: +0.51% | Improved |
| Add parameter diversity loss to logits-only mutual learning (ablation) | Bayesian mutual learning, ResNet20 initialized from deterministic-model mean · CIFAR-100 | top-1 accuracy: +0.61% | Improved |
| Add feature diversity loss to logits-only mutual learning (ablation) | Bayesian mutual learning, ResNet20 initialized from scratch · CIFAR-100 | top-1 accuracy: +0.3% | Improved |
| Add feature diversity loss to logits-only mutual learning (ablation) | Bayesian mutual learning, ResNet20 initialized from deterministic-model mean · CIFAR-100 | top-1 accuracy: +0.43% | Improved |
| Use both parameter and feature diversity losses rather than logits-only training (ablation) | Bayesian mutual learning · CIFAR-100 | top-1 accuracy | Improved |
| Use fused-feature distribution diversity rather than directly maximizing feature distance (ablation) | Bayesian mutual learning · CIFAR-100 | top-1 accuracy | Improved |
Closest papers by meaning
- Natural-Parameter Networks: A Class of Probabilistic Neural Networks (NeurIPS 2016 · similarity 0.87)
- Encoding the Latent Posterior of Bayesian Neural Networks for Uncertainty Quantification (TPAMI 2023 · similarity 0.87)
- DIBS: Diversity Inducing Information Bottleneck in Model Ensembles (AAAI 2021 · similarity 0.87)
- Make Me a BNN: A Simple Strategy for Estimating Bayesian Uncertainty from Pre-trained Models (CVPR 2024 · similarity 0.86)
- A Study of Bayesian Neural Network Surrogates for Bayesian Optimization (ICLR 2024 · similarity 0.86)
- Parameter Expanded Stochastic Gradient Markov Chain Monte Carlo (ICLR 2025 · similarity 0.85)
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.
