Abstract:We study the problem of bounding the posterior distribution of discrete probabilistic programs with unbounded support, loops, and conditioning. Loops pose the main difficulty in this setting: even if exact Bayesian inference is possible, the state of the art requires user-provided loop invariant templates. By contrast, we aim to find guaranteed bounds, which sandwich the true distribution. They are fully automated, applicable to more programs and provide more provable guarantees than approximate sampling-based inference. Since lower bounds can be obtained by unrolling loops, the main challenge is upper bounds, and we attack it in two ways. The first is called residual mass semantics, which is a flat bound based on the residual probability mass of a loop. The approach is simple, efficient, and has provable guarantees. The main novelty of our work is the second approach, called geometric bound semantics. It operates on a novel family of distributions, called eventually geometric distributions (EGDs), and can bound the distribution of loops with a new form of loop invariants called contraction invariants. The invariant synthesis problem reduces to a system of polynomial inequality constraints, which is a decidable problem with automated solvers. If a solution exists, it yields an exponentially decreasing bound on the whole distribution, and can therefore bound moments and tail asymptotics as well, not just probabilities as in the first approach. Both semantics enjoy desirable theoretical properties. In particular, we prove soundness and convergence, i.e. the bounds converge to the exact posterior as loops are unrolled further. On the practical side, we describe Diabolo, a fully-automated implementation of both semantics, and evaluate them on a variety of benchmarks from the literature, demonstrating their general applicability and the utility of the resulting bounds.

Data-driven invariant learning for probabilistic programs

Learning Likely Invariants to Explain Why a Program Fails

Data-Driven Template-Free Invariant Generation

Probabilistic Program Verification Via Inductive Synthesis of Inductive Invariants.

Probabilistic Conditional System Invariant Generation with Bayesian Inference

A Deductive Verification Infrastructure for Probabilistic Programs

Using Dynamic Analysis to Generate Disjunctive Invariants

Provably Invariant Learning Without Domain Information.

Learning Probabilistic Logic Programs in Continuous Domains

A novel data-driven approach on inferring loop invariants for C programs

A Framework for Safe Probabilistic Invariance Verification of Stochastic Dynamical Systems

Piecewise Linear Expectation Analysis via $k$-Induction for Probabilistic Programs

Beyond the elementary representations of program invariants over algebraic data types

A Novel Data-Driven Approach for Generating Verified Loop Invariants *

Probabilistic Invariant Learning with Randomized Linear Classifiers

A robust assessment for invariant representations

Invariant Probabilistic Prediction

Guaranteed Bounds on Posterior Distributions of Discrete Probabilistic Programs with Loops

Latticed K-Induction with an Application to Probabilistic Programs

Conformal Inference for Invariant Risk Minimization

Integrated Latent Heterogeneity and Invariance Learning in Kernel Space