Research map / Optimization / Optimizers
Interacting Contour Stochastic Gradient Langevin Dynamics
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 |
|---|---|---|---|
| Average the random-field updates across parallel chains instead of adapting each chain independently | CSGLD · Mushroom contextual bandit | cumulative regret | Helped |
| Use interacting chains and the scalable random-field function | CSGLD · Multimodal distribution simulation | mode-weight approximation | Helped |
| Use interacting contour sampling instead of SGD | MNIST digits 0–4 · MNIST mode exploration | mode exploration | Helped |
| Use interacting contour sampling instead of SGLD | MNIST digits 0–4 · MNIST mode exploration | mode exploration | Helped |
| Use ICSGHMC×P4 instead of reSGHMC×P4 | CIFAR100 uncertainty estimation · CIFAR100 test, ResNet20 and ResNet32 | ACC (%): 76.15$\pm$0.16 -> 76.34$\pm$0.15; 78.57$\pm$0.07 -> 78.72$\pm$0.16 | Helped |
| Use ICSGHMC×P4 instead of M-SGD×P4 | CIFAR100 uncertainty estimation · SVHN, ResNet20 and ResNet32 | Brier (‰): 2.58$\pm$0.08 -> 2.54$\pm$0.14; 2.77$\pm$0.15 -> 2.76$\pm$0.15 | Mixed |
| Use ICSGHMC×P4 instead of M-SGD×P4 | CIFAR100 uncertainty estimation · SVHN, ResNet56 and WRN-16-8 | Brier (‰): 2.86$\pm$0.08 -> 2.88$\pm$0.09; 1.77$\pm$0.06 -> 1.83$\pm$0.06 | Hurt |
| Use interacting chains rather than a single chain with equivalent computational budget | CSGLD · Theoretical stochastic-approximation analysis | asymptotic variance | Helped |
| Use single-chain CSGLD×T4 instead of pSGLD×P4 | Mushroom contextual bandit · Mushroom contextual bandit | cumulative regret | Hurt |
| Use cycSWAG×T4 instead of ICSGHMC×P4 | CIFAR100 uncertainty estimation · CIFAR100 test, WRN-16-8 | NLL: 6359$\pm$14 -> 6338$\pm$36 | Hurt |
Closest papers by meaning
- A Contour Stochastic Gradient Langevin Dynamics Algorithm for Simulations of Multi-modal Distributions (NeurIPS 2020 · similarity 0.93)
- Accelerating the diffusion-based ensemble sampling by non-reversible dynamics (ICML 2020 · similarity 0.89)
- Constrained Exploration via Reflected Replica Exchange Stochastic Gradient Langevin Dynamics (ICML 2024 · similarity 0.87)
- Faster Sampling via Stochastic Gradient Proximal Sampler (ICML 2024 · similarity 0.87)
- Provable Convergence and Limitations of Geometric Tempering for Langevin Dynamics (ICLR 2025 · similarity 0.86)
- Sequential Controlled Langevin Diffusions (ICLR 2025 · 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.