Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale

Abstract

Probabilistic programming languages (PPLs) are receiving widespread attention for performing Bayesian inference in complex generative models. However, applications to science remain limited because of the impracticability of rewriting complex scientific simulators 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 provides 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 minibatch size of 128k: achieving a performance of 450 Tflop/s through enhancements to PyTorch. We demonstrate a Large Hadron Collider (LHC) use-case with the C++ Sherpa simulator and achieve the largest-scale posterior inference in a Turing-complete PPL.

14 pages, 8 figures

References in corpus (10)

• Event generation with SHERPA 1.1
• The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo
• Performance of the ATLAS Trigger System in 2015
• Fast inference of deep neural networks in FPGAs for particle physics
• Search for the Standard Model Higgs boson produced in association with top quarks and decaying into $b\bar{b}$ in pp collisions at $\sqrt{s}$ = 8 TeV with the ATLAS detector
• Semantics for probabilistic programming: higher-order functions, continuous distributions, and soft constraints
• Approximate Bayesian Computation for Forward Modeling in Cosmology
• Reconstruction of hadronic decay products of tau leptons with the ATLAS experiment
• Accelerating recurrent neural network training using sequence bucketing and multi-GPU data parallelization
• Scaling GRPC Tensorflow on 512 nodes of Cori Supercomputer

Cited by in corpus (10)

• The frontier of simulation-based inference
• Technology Readiness Levels for Machine Learning Systems
• Rare and Different: Anomaly Scores from a combination of likelihood and out-of-distribution models to detect new physics at the LHC
• A Probabilistic Machine Learning Approach to Scheduling Parallel Loops with Bayesian Optimization
• Reframing Jet Physics with New Computational Methods
• Bayesian Neural Networks at Scale: A Performance Analysis and Pruning Study
• Planning as Inference in Epidemiological Models
• Statically Bounded-Memory Delayed Sampling for Probabilistic Streams
• stanhf: HistFactory models in the probabilistic programming language Stan
• High Throughput Training of Deep Surrogates from Large Ensemble Runs