VBF Experiments, April 2026

Amortizing Quadrature Filters Without Losing Calibration

A strict mixture filter learned from deterministic Power-EP teachers improved some nonlinear alias cases, but exposed a hard alias-mass calibration tradeoff.

Apr 30, 2026 ai-research, variational-filtering, nonlinear-filtering, assumed-density-filtering, expectation-propagation

A Tour Of Learned And Reference-Free Bayesian Filters

A long-form guide to the Kalman, ELBO, distillation, IWAE, FIVO, ADF, and Power-EP filtering experiments in ml-examples.

Apr 30, 2026 ai-research, bayesian-filtering, variational-filtering, nonlinear-filtering, assumed-density-filtering

Reference-Free Quadrature Filters For The Sine Benchmark

Deterministic quadrature ADF and Power-EP baselines showed that much of the nonlinear filtering gap was algorithmic, not just amortization.

Apr 30, 2026 ai-research, nonlinear-filtering, assumed-density-filtering, expectation-propagation

When Mixtures Beat Local ELBO In Nonlinear Filtering

Small strict mixture filters with IWAE and FIVO-style objectives closed much of the nonlinear calibration gap while staying reference-free.

Apr 30, 2026 ai-research, variational-filtering, nonlinear-filtering, iwae, particle-filtering

Repairing a Nonlinear Strict Filter Without Reference Targets

The nonlinear sine-observation benchmark exposed ELBO under-dispersion, then a joint ELBO, predictive-y, and masked-y objective partially repaired it.

Apr 30, 2026 ai-research, variational-filtering, nonlinear-filtering, elbo

Variational Filtering, Rebuilt From the Linear Case

A scalar linear-Gaussian benchmark made the VBF edge-factor implementation auditable before moving to nonlinear filtering.

Apr 30, 2026 ai-research, variational-filtering, state-space-models, calibration