Trace types and denotational semantics for sound programmable inference in probabilistic languages

Probabilistic programming languages have the potential to make probabilistic modeling and inference easier to use in practice, but only if inference is sufficiently fast and accurate for real applications. Thus far, this has only been possible for domain-specific languages that focus on a restricted class of models and inference algorithms. This paper proposes a design for a probabilistic programming language called Gen, embedded in Julia, that aims to be sufficiently expressive and performant for general-purpose use. The language provides constructs for automatically generating optimized implementations of custom inference tactics based on static analysis of the target probabilistic model. This paper informally describes a language design for Gen, and shows that Gen is more expressive than Stan, a widely used language for hierarchical Bayesian modeling. A first benchmark shows that a prototype implementation of Gen can be as fast as Stan, only ∼1.4x slower than a hand-coded sampler in Julia, and ∼7,500x faster than Venture, one of the only other probabilistic languages with support for custom inference.

Probabilistic programming languages aim to separate the user-specified model from the inference runtime system, allowing users to focus on model design and leave inference to the compiler and runtime systems. However, the structure of the model can affect the inference efficiency, something that is often utilized when implementing Bayesian inference solutions without probabilistic programming. In this paper, we consider a specific class of important models within phylogenetics, which are used for tree inference. Algorithms utilizing the structures of trees in such models have been used for a long time within phylogenetics, outside the scope of probabilistic programming. In particular, the use of belief propagation can significantly speed up probabilistic tree inference when combined with other approximate inferences, such as MCMC. However, there is an open problem of how to incorporate both belief-propagation (marginalization) of a subset of latent variables in typed universal probabilistic programming languages and still be able to apply approximate inference for the rest of the variables. In this paper, we propose a solution to this problem by using an approach where a user only needs to mark which latent variables in the program should be marginalized, and the rest of the belief propagation is handled automatically. We call this approach annotated automatic pruning, acknowledging the original terminology of pruning from the phylogenetic literature. We evaluate our approach using non-trivial models for tree inference, assessing performance, and testing correctness by comparing to MrBayes as the reference implementation.

Probabilistic Programming Languages (PPLs) allow users to encode statistical inference problems and automatically apply an inference algorithm to solve them. Popular inference algorithms for PPLs, such as sequential Monte Carlo (SMC) and Markov chain Monte Carlo (MCMC), are built around checkpoints—relevant events for the inference algorithm during the execution of a probabilistic program. Deciding the location of checkpoints is, in current PPLs, not done optimally. To solve this problem, we present a static analysis technique that automatically determines checkpoints in programs, relieving PPL users of this task. The analysis identifies a set of checkpoints that execute in the same order in every program run—they are aligned. We formalize alignment, prove the correctness of the analysis, and implement the analysis as part of the higher-order functional PPL Miking CorePPL. By utilizing the alignment analysis, we design two novel inference algorithm variants: aligned SMC and aligned lightweight MCMC. We show, through real-world experiments, that they significantly improve inference execution time and accuracy compared to standard PPL versions of SMC and MCMC.

Probabilistic programming languages (PPLs) are an expressive means for creating and reasoning about probabilistic models. Unfortunately hybrid probabilistic programs that involve both continuous and discrete structures are not well supported by today’s PPLs. In this paper we develop a new approximate inference algorithm for hybrid probabilistic programs that first discretizes the continuous distributions and then performs discrete inference on the resulting program. The key novelty is a form of discretization that we call bit blasting , which uses a binary representation of numbers such that a domain of 2 b discretized points can be succinctly represented as a discrete probabilistic program over poly b Boolean random variables. Surprisingly, we prove that many common continuous distributions can be bit blasted in a manner that incurs no loss of accuracy over an explicit discretization and supports efficient probabilistic inference. We have built a probabilistic programming system for hybrid programs called HyBit , which employs bit blasting followed by discrete probabilistic inference. We empirically demonstrate the benefits of our approach over existing sampling-based and symbolic inference approaches

Probabilistic programming refers to the idea of using standard programming constructs for specifying probabilistic models from machine learning and statistics, and employing generic inference algorithms for answering various queries on these models, such as posterior inference and estimation of model evidence. Although this idea itself is not new and was, in fact, explored by several programming-language and statistics researchers in the early 2000, it is only in the last few years that probabilistic programming has gained a large amount of attention among researchers in machine learning and programming languages, and that expressive and efficient probabilistic programming systems (such as Anglican, Church, Figaro, Infer.net, PSI, PyMC, Stan, and Venture) started to appear and have been taken up by a nontrivial number of users. The primary goal of my talk is to introduce probabilistic programming to the CONCUR/QUEST/FORMATS audience. At the end of my talk, I want the audience to understand basic results and techniques in probabilistic programming and to feel that these results and techniques are relevant or at least related to what she or he studies, although they typically come from foreign research areas, such as machine learning and statistics. My talk will contain both technical materials and lessons that I learnt from my machine-learning colleagues in Oxford, who are developing a highly-expressive higher-order probabilistic programming language, called Anglican. It will also include my work on the denotational semantics of higher-order probabilistic programming languages and their inference algorithms, which are jointly pursued with colleagues in Cambridge, Edinburgh, Oxford and Tubingen.

In Bayesian probabilistic programming, a central problem is to estimate the normalised posterior distribution (NPD) of a probabilistic program with conditioning via score (a.k.a. observe ) statements. Most previous approaches address this problem by Markov Chain Monte Carlo and variational inference, and therefore could not generate guaranteed outcomes within a finite time limit. Moreover, existing methods for exact inference either impose syntactic restrictions or cannot guarantee successful inference in general. In this work, we propose a novel automated approach to derive guaranteed bounds for NPD via polynomial solving. We first establish a fixed-point theorem for the wide class of score-at-end Bayesian probabilistic programs that terminate almost-surely and have a single bounded score statement at program termination. Then, we propose a multiplicative variant of Optional Stopping Theorem (OST) to address score-recursive Bayesian programs where score statements with weights greater than one could appear inside a loop. Bayesian nonparametric models, enjoying a renaissance in statistics and machine learning, can be represented by score-recursive Bayesian programs and are difficult to handle due to an integrability issue. Finally, we use polynomial solving to implement our fixed-point theorem and OST variant. To improve the accuracy of the polynomial solving, we further propose a truncation operation and the synthesis of multiple bounds over various program inputs. Our approach can handle Bayesian probabilistic programs with unbounded while loops and continuous distributions with infinite supports. Experiments over a wide range of benchmarks show that compared with the most relevant approach (Beutner et al. , PLDI 2022) for guaranteed NPD analysis via recursion unrolling, our approach is more time efficient and derives comparable or even tighter NPD bounds. Furthermore, our approach can handle score-recursive programs which previous approaches could not.

Applications of graphical models often require the use of approximate inference, such as sequential importance sampling (SIS), for estimation of the model distribution given partial evidence, i.e., the target distribution. However, when SIS proposal and target distributions are dissimilar, such procedures lead to biased estimates or require a prohibitive number of samples. We introduce ReBaSIS, a method that better approximates the target distribution by sampling variable by variable from existing importance samplers and accepting or rejecting each proposed assignment in the sequence: a choice made based on anticipating upcoming evidence. We relate the per-variable proposal and model distributions by expected weight ratios of sequence completions and show that we can learn accurate models of optimal acceptance probabilities from local samples. In a continuous-time domain, our method improves upon previous importance samplers by transforming an SIS problem into a machine learning one.

Publication in the conference proceedings of EUSIPCO, Toulouse, France, 2002

Sequential importance sampling for structural reliability analysis

Probabilistic programming languages (PPLs) are at the interface between statistics and the theory of programming languages. PPLs formulate statistical models as stochastic programs that enable automatic inference algorithms and optimization. Pyro [1] and its sibling NumPyro [2] are PPLs built on top of the deep learning frameworks PyTorch [3] and Jax [4], respectively. Both PPLs provide simple, highly similar interfaces for inference using efficient implementations of Hamiltonian Monte Carlo (HMC), the No-U-Turn Sampler (NUTS), and Stochastic Variational Inference (SVI). They automatically generate variational distributions from a model, automatically enumerate discrete variables, and support formulating deep probabilistic models such as variational autoencoders and deep Markov models. The Sine von Mises distribution and its skewed variant are toroidal distributions relevant to protein bioinformatics. They provide a natural way to model the dihedral angles of protein structures, which is important in protein structure prediction, simulation and analysis. We present efficient implementations of the Sine von Mises distribution and its skewing in Pyro and NumPyro, and devise a simulation method that increases efficiency with several orders of magnitude when using parallel hardware (i.e., modern CPUs, GPUs, and TPUs). We demonstrate the use of the skewed Sine von Mises distribution by modeling dihedral angles of proteins using a Bayesian mixture model inferred using NUTS, exploiting NumPyro's facilities for automatic enumeration [5].

Coalescent-based Bayesian Markov chain Monte Carlo (MCMC) inference generates estimates of evolutionary parameters and their posterior probability distributions. As the number of sequences increases, the length of time taken to complete an MCMC analysis increases as well. Here, we investigate an approach to distribute the MCMC analysis across a cluster of computers. To do this, we use bootstrapped topologies as fixed genealogies, perform a single MCMC analysis on each genealogy without topological rearrangements, and pool the results across all MCMC analyses. We show, through simulations, that although the standard MCMC performs better than the bootstrap-MCMC at estimating the effective population size (scaled by mutation rate), the bootstrap-MCMC returns better estimates of growth rates. Additionally, we find that our bootstrap-MCMC analyses are, on average, 37 times faster for equivalent effective sample sizes.

Satellite data provides essential insights into the spatiotemporal distribution of CO2 concentrations. However, many atmospheric inverse models fail to adequately incorporate the spatial and temporal correlations inherent in satellite observations and often lack rigorous methods for estimating parameters like spatial length scales. We introduce an inference model that processes the spatiotemporal covariance in satellite data and estimates hyperparameters such as covariance length scales. Our approach uses the Gaussian process (GP) machine learning (ML) and modern probabilistic programming languages (PPLs) to perform atmospheric inversions of emissions from satellite data. We develop a GP ML inversion system based on modern PPLs and the GEOS-Chem chemical transport model, simulating atmospheric CO2 concentrations corresponding to the Orbiting Carbon Observatory-2/3 (OCO-2/3) data for July 2020. In our supervised learning framework, we treat the GEOS-Chem simulated data set as the target, with predictors derived by scaling the target with sector-specific factors hidden from the GP machine. Our results show that the GP model, combined with GPU-enabled PPLs, effectively retrieves true emission scaling factors and infers noise levels concealed within the data. This suggests that our method could be applied over larger areas with more complex covariance structures, enabling comprehensive analysis of the spatiotemporal patterns observed in OCO-2/3 and similar satellite data sets.

Probabilistic programming languages (PPLs) allow programmers to construct statistical models and then simulate data or perform inference over them. Many PPLs restrict models to a particular instance of simulation or inference, limiting their reusability. In other PPLs, models are not readily composable. Using Haskell as the host language, we present an embedded domain specific language based on algebraic effects, where probabilistic models are modular, first-class, and reusable for both simulation and inference. We also demonstrate how simulation and inference can be expressed naturally as composable program transformations using algebraic effect handlers.

This system description paper introduces the probabilistic programming language Hakaru10, for expressing, and performing inference on (general) graphical models. The language supports discrete and continuous distributions, mixture distributions and conditioning. Hakaru10 is a DSL embedded in Haskell and supports Monte-Carlo Markov Chain (MCMC) inference.

Opening doors is an essential task that a robot should perform. In this paper, we propose a logical approach to predict the action of opening doors, together with the action point where the action should be performed. The input of our system is a pair of bounding boxes of the door and door handle, together with background knowledge in the form of logical rules. Learning and inference are performed with the probabilistic programming language ProbLog. We evaluate our approach on a doors dataset and we obtain encouraging results. Additionally, a comparison to a propositional decision tree shows the benefits of using a probabilistic programming language such as ProbLog.