skip to main content


Title: Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale
Probabilistic programming languages (PPLs) are receiving wide- spread attention for performing Bayesian inference in complex generative models. However, applications to science remain limited because of the impracticability of rewriting complex scientific simu- lators in a PPL, the computational cost of inference, and the lack of scalable implementations. To address these, we present a novel PPL framework that couples directly to existing scientific simulators through a cross-platform probabilistic execution protocol and pro- vides Markov chain Monte Carlo (MCMC) and deep-learning-based inference compilation (IC) engines for tractable inference. To guide IC inference, we perform distributed training of a dynamic 3DCNN– LSTM architecture with a PyTorch-MPI-based framework on 1,024 32-core CPU nodes of the Cori supercomputer with a global mini- batch size of 128k: achieving a performance of 450 Tflop/s through enhancements to PyTorch. We demonstrate a Large Hadron Col- lider (LHC) use-case with the C++ Sherpa simulator and achieve the largest-scale posterior inference in a Turing-complete PPL.  more » « less
Award ID(s):
1836650
PAR ID:
10169392
Author(s) / Creator(s):
Date Published:
Journal Name:
SC '19: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Modern scientific workflows couple simulations with AI-powered analytics by frequently exchanging data to accelerate time-to-science to reduce the complexity of the simulation planes. However, this data exchange is limited in performance and portability due to a lack of support for scientific data formats in AI frameworks. We need a cohesive mechanism to effectively integrate at scale complex scientific data formats such as HDF5, PnetCDF, ADIOS2, GNCF, and Silo into popular AI frameworks such as TensorFlow, PyTorch, and Caffe. To this end, we designed Stimulus, a data management library for ingesting scientific data effectively into the popular AI frameworks. We utilize the StimOps functions along with StimPack abstraction to enable the integration of scientific data formats with any AI framework. The evaluations show that Stimulus outperforms several large-scale applications with different use-cases such as Cosmic Tagger (consuming HDF5 dataset in PyTorch), Distributed FFN (consuming HDF5 dataset in TensorFlow), and CosmoFlow (converting HDF5 into TFRecord and then consuming that in TensorFlow) by 5.3 x, 2.9 x, and 1.9 x respectively with ideal I/O scalability up to 768 GPUs on the Summit supercomputer. Through Stimulus, we can portably extend existing popular AI frameworks to cohesively support any complex scientific data format and efficiently scale the applications on large-scale supercomputers. 
    more » « less
  2. Abstract

    Inference is crucial in modern astronomical research, where hidden astrophysical features and patterns are often estimated from indirect and noisy measurements. Inferring the posterior of hidden features, conditioned on the observed measurements, is essential for understanding the uncertainty of results and downstream scientific interpretations. Traditional approaches for posterior estimation include sampling-based methods and variational inference (VI). However, sampling-based methods are typically slow for high-dimensional inverse problems, while VI often lacks estimation accuracy. In this paper, we proposeα-deep probabilistic inference, a deep learning framework that first learns an approximate posterior usingα-divergence VI paired with a generative neural network, and then produces more accurate posterior samples through importance reweighting of the network samples. It inherits strengths from both sampling and VI methods: it is fast, accurate, and more scalable to high-dimensional problems than conventional sampling-based approaches. We apply our approach to two high-impact astronomical inference problems using real data: exoplanet astrometry and black hole feature extraction.

     
    more » « less
  3. Hamiltonian Monte Carlo (HMC) is a powerful algorithm to sample latent variables from Bayesian models. The advent of probabilistic programming languages (PPLs) frees users from writing inference algorithms and lets users focus on modeling. However, many models are difficult for HMC to solve directly, and often require tricks like model reparameterization. We are motivated by the fact that many of those models could be simplified by marginalization. We propose to use automatic marginalization as part of the sampling process using HMC in a graphical model extracted from a PPL, which substantially improves sampling from real-world hierarchical models. 
    more » « less
  4. This paper presents ProbCompCert, a compiler for a subset of the Stan probabilistic programming language (PPL), in which several key compiler passes have been formally verified using the Coq proof assistant. Because of the probabilistic nature of PPLs, bugs in their compilers can be difficult to detect and fix, making verification an interesting possibility. However, proving correctness of PPL compilation requires new techniques because certain transformations performed by compilers for PPLs are quite different from other kinds of languages. This paper describes techniques for verifying such transformations and their application in ProbCompCert. In the course of verifying ProbCompCert, we found an error in the Stan language reference manual related to the semantics and implementation of a key language construct. 
    more » « less
  5. Cone Penetration Test (CPT) based simplified liquefaction evaluation procedures offer significant advantages over variants based on alternative in-situ test indices. However, the main drawback of CPT-based variants is that soil samples are typically not recovered during CPT sounding, and as a result, soils are often not characterized directly or tested further in the laboratory. The need thus arises to infer soil properties from CPT indices without physical confirmation. Utilizing an unprecedented database of field and laboratory test data from Christchurch, New Zealand, this study proposes deterministic and probabilistic correlations relating the soil behavior type index (Ic) to liquefaction susceptibility as defined by published criteria. The findings show that Ic can be used with reasonable accuracy as a proxy for laboratory index-test based criteria for assessing liquefaction susceptibility. Additionally, to assess liquefaction hazards in a fully probabilistic manner, the uncertainty of Ic for defining liquefaction susceptibility can be accounted for using an approach similar to that used to create fragility functions for performance-based earthquake engineering. The approaches used herein to develop deterministic and probabilistic Ic relationships for liquefaction susceptibility are not limited to soils from New Zealand, but rather, can be applied worldwide. 
    more » « less