Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Replica-Exchange Nos\'e-Hoover Dynamics for Bayesian Learning on Large Datasets
Rui Luo, Qiang Zhang, Yaodong Yang, Jun Wang
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 |
|---|---|---|---|
| Use RENHD instead of TACT-HMC | LSTM RNN Bayesian sampler · Fashion-MNIST, 0% permuted labels | accuracy: 89.74 \pm 0.13\,\% -> 90.87 \pm 0.12\,\% | Improved |
| Use RENHD instead of TACT-HMC | LSTM RNN Bayesian sampler · Fashion-MNIST, 20% permuted labels | accuracy: 88.83 \pm 0.17\,\% -> 89.45 \pm 0.17\,\% | Improved |
| Use RENHD instead of TACT-HMC | LSTM RNN Bayesian sampler · Fashion-MNIST, 30% permuted labels | accuracy: 88.78 \pm 0.17\,\% -> 89.06 \pm 0.16\,\% | Improved |
| Use RENHD instead of TACT-HMC | ResNet Bayesian sampler · CIFAR-10, 0% permuted labels | accuracy: 88.01 \pm 0.13\,\% -> 88.41 \pm 0.12\,\% | Improved |
| Use RENHD instead of TACT-HMC | ResNet Bayesian sampler · CIFAR-10, 20% permuted labels | accuracy: 82.28 \pm 0.13\,\% -> 84.48 \pm 0.14\,\% | Improved |
| Use RENHD instead of TACT-HMC | ResNet Bayesian sampler · CIFAR-10, 30% permuted labels | accuracy: 79.43 \pm 0.12\,\% -> 82.65 \pm 0.13\,\% | Improved |
| Use RENHD instead of TACT-HMC | Bayesian sampler · Real-image experiments, same execution time | unbiased multimodal samples: 4 times | Improved |
| Use the analytic compensation distribution rather than the numerical solution | Replica-exchange acceptance protocol · Sampling | sampling speed: 20\times | Improved |
Closest papers by meaning
- Non-convex Learning via Replica Exchange Stochastic Gradient MCMC (ICML 2020 · similarity 0.86)
- Constrained Exploration via Reflected Replica Exchange Stochastic Gradient Langevin Dynamics (ICML 2024 · similarity 0.85)
- Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics (ICLR 2025 · similarity 0.84)
- Diffusion Models are Secretly Exchangeable: Parallelizing DDPMs via Auto Speculation (ICML 2025 · similarity 0.83)
- BARNN: A Bayesian Autoregressive and Recurrent Neural Network (ICML 2025 · similarity 0.83)
- Stochastic Forward–Backward Deconvolution: Training Diffusion Models with Finite Noisy Datasets (ICML 2025 · similarity 0.83)
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.
