Sampling ideas

Diffusion step schedules, solvers, guidance and stopping rules that cut function evaluations or improve sample quality.

2 ideas found

Category: Sampling ML: Optimizer Math: Statistics Clear all

✗ Failed on benchmark 2026

Cramer-Rao Temperature Controller

Use the generalized Cramér–Rao relation to adapt the inverse-temperature or noise schedule of an energy-based sampler, diffusion sampler, or stochastic optimizer. The controller limits each temperature change according to the measured energy variance and Fisher information, preventing uncontrolled changes in the sampled energy distribution while allowing larger steps in insensitive regions.

Useful8/10

Difficulty4/10

Novelty7/10

Sampling Stability Speedup Sample efficiency

Paper: Cramer-Rao Inequality Generalizes the Equilibrium Energy Fluctuation-Response Relation to Nonequilibrium Steady States arXiv:2608.23455

✗ Mechanism failed 2026

Controlled Stationary Hyperparameter Sweep

Replace many independently equilibrated SGLD runs at different hyperparameters with one controlled sweep in which an auxiliary drift transports particles through the stationary distributions indexed by the swept parameter. Estimate the response of loss, predictions, uncertainty, or weight observables using covariance with the stationary generalized-potential derivative instead of finite differences between separate runs.

Useful7/10

Difficulty7/10

Novelty7/10

Sampling Speedup Stability Sample efficiency

Paper: Shortcuts to Parameter Sweeps arXiv:2608.12154