Proposal for a New LogDensity Function Interface

[sunxd3]

TuringLang/DynamicPPL.jl#691

The goal is to design a flexible, user-friendly interface for log density functions that can handle various model operations, especially in higher-order contexts like Gibbs sampling and Bayesian workflows.

Evaluation functions:

1. evaluate

Query functions:

1. is_parametric(model)
2. dimension(model) (only defined when is_parametric(model) == true)
3. is_conditioned(model)
4. is_fixed(model)
5. logjoint(model, params)
6. loglikelihood(model, params)
7. logprior(model, params)

where params can be Vector, NamedTuple, Dict, etc.

Transformation functions:

1. condition(model, conditioned_vars)
2. fix(model, fixed_vars)
3. factorize(model, variables_in_the_factor)

condition and factor are similar, but factorize effectively generates a sub-model.

Higher-order functions:

1. generated_quantities(model, sample, [, expr]) or generated_quantities(model, sample, f, args...)
   1. generated_quantities computes things from the sampling result.
   2. In DynamicPPL, this is the model's return value. For more flexibility, we should allow passing an expression or function. (Currently, users can rewrite the model definition to achieve this in DynamicPPL, but with limitations. We want to make this more generic.)
   3. rand is a special case of generated_quantities (when no sample is passed).
2. predict(model, sample)
3. simulation_based_calibration
   Runs the full model + inference pipeline N times to check calibration:
   1. Draw true parameters θ₀ ∼ p(θ).
   2. Generate synthetic data y ∼ p(y | θ₀).
   3. Run inference(condition(model, y)) to obtain posterior draws {θᵢ}.
   4. Compute the rank of summary(θ₀) within summary.(θᵢ).
   5. Aggregate ranks over replications and apply discrepancy (default KS) to return an SBCResult with ranks, histogram, p-value, etc.

generated_quantities can be implemented by fixing the model on sample and calling evaluate.
predict can be implemented by unconditioning the model on data, fixing it on sample, and calling evaluate.

[sunxd3]

tagging @penelopeysm for a look

[yebai]

for reference: TuringLang/DynamicPPL.jl#900