Research map / Probabilistic and scientific modeling / Variational and generative sampling
Geometrically Coupled Monte Carlo Sampling
Mark Rowland, Krzysztof M Choromanski, François Chalus, Aldo Pacchiano, Tamas Sarlos, Richard E Turner, Adrian Weller
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 orthogonal samples instead of i.i.d. Gaussian samples | ES gradient estimation · Minitaur locomotion, m << d | reward / walkable policy (R > 10) | Improved |
| couple lengths of antithetic pairs (Algorithm 1) with orthogonal samples (ablation) | MCGaussianOrthogonal antithetic · Minitaur locomotion, m=8 | walkable policy found | Improved |
| use randomized Halton QMC sequences instead of Gaussian/orthogonal samples | ES gradient estimation · Minitaur locomotion, m=96 | walkable policy found | Hurt |
| use orthogonal or antithetic length-coupled samples instead of i.i.d. | Gaussian sample sequences · star discrepancy of 40-sample sequences | D* discrepancy | Improved |
| use GCMC sampling instead of i.i.d. samples | VAE ELBO estimation · MNIST VAE | train/test ELBO after 50 epochs | Improved |
| RQMC vs coupled antithetic samples for low discrepancy | sampling methods · star discrepancy | D* discrepancy | Mixed |
Closest papers by meaning
- Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics (ICLR 2025 · similarity 0.87)
- Transport meets Variational Inference: Controlled Monte Carlo Diffusions (ICLR 2024 · similarity 0.86)
- Generative diffusion for perceptron problems: statistical physics analysis and efficient algorithms (NeurIPS 2025 · similarity 0.86)
- Provable and Practical: Efficient Exploration in Reinforcement Learning via Langevin Monte Carlo (ICLR 2024 · similarity 0.86)
- NETS: A Non-equilibrium Transport Sampler (ICML 2025 · similarity 0.86)
- Improved Lower Bounds for First-order Stochastic Non-convex Optimization under Markov Sampling (ICML 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.
