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.
Award ID(s):
Publication Date:
Journal Name:
SC '19: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
Sponsoring Org:
National Science Foundation
More Like this
  1. 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 propose α -deep probabilistic inference, a deep learning framework that first learns an approximate posterior using α -divergence VI paired with a generativemore »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.« less
  2. A cornerstone of theoretical neuroscience is the circuit model: a system of equations that captures a hypothesized neural mechanism. Such models are valuable when they give rise to an experimentally observed phenomenon -- whether behavioral or a pattern of neural activity -- and thus can offer insights into neural computation. The operation of these circuits, like all models, critically depends on the choice of model parameters. A key step is then to identify the model parameters consistent with observed phenomena: to solve the inverse problem. In this work, we present a novel technique, emergent property inference (EPI), that brings themore »modern probabilistic modeling toolkit to theoretical neuroscience. When theorizing circuit models, theoreticians predominantly focus on reproducing computational properties rather than a particular dataset. Our method uses deep neural networks to learn parameter distributions with these computational properties. This methodology is introduced through a motivational example of parameter inference in the stomatogastric ganglion. EPI is then shown to allow precise control over the behavior of inferred parameters and to scale in parameter dimension better than alternative techniques. In the remainder of this work, we present novel theoretical findings in models of primary visual cortex and superior colliculus, which were gained through the examination of complex parametric structure captured by EPI. Beyond its scientific contribution, this work illustrates the variety of analyses possible once deep learning is harnessed towards solving theoretical inverse problems.« less
  3. The theoretical foundations of Big Data Science are not fully developed, yet. This study proposes a new scalable framework for Big Data representation, high-throughput analytics (variable selection and noise reduction), and model-free inference. Specifically, we explore the core principles of distribution-free and model-agnostic methods for scientific inference based on Big Data sets. Compressive Big Data analytics (CBDA) iteratively generates random (sub)samples from a big and complex dataset. This subsampling with replacement is conducted on the feature and case levels and results in samples that are not necessarily consistent or congruent across iterations. The approach relies on an ensemble predictor wheremore »established model-based or model-free inference techniques are iteratively applied to preprocessed and harmonized samples. Repeating the subsampling and prediction steps many times, yields derived likelihoods, probabilities, or parameter estimates, which can be used to assess the algorithm reliability and accuracy of findings via bootstrapping methods, or to extract important features via controlled variable selection. CBDA provides a scalable algorithm for addressing some of the challenges associated with handling complex, incongruent, incomplete and multi-source data and analytics challenges. Albeit not fully developed yet, a CBDA mathematical framework will enable the study of the ergodic properties and the asymptotics of the specific statistical inference approaches via CBDA. We implemented the high-throughput CBDA method using pure R as well as via the graphical pipeline environment. To validate the technique, we used several simulated datasets as well as a real neuroimaging-genetics of Alzheimer’s disease case-study. The CBDA approach may be customized to provide generic representation of complex multimodal datasets and to provide stable scientific inference for large, incomplete, and multisource datasets.« less
  4. Real-world tasks often exhibit a compositional structure that contains a sequence of simpler sub-tasks. For instance, opening a door requires reaching, grasping, rotating, and pulling the door knob. Such compositional tasks require an agent to reason about the sub-task at hand while orchestrating global behavior accordingly. This can be cast as an online task inference problem, where the current task identity, represented by a context variable, is estimated from the agent’s past experiences with probabilistic inference. Previous approaches have employed simple latent distributions, e.g., Gaussian, to model a single context for the entire task. However, this formulation lacks the expressivenessmore »to capture the composition and transition of the sub-tasks. We propose a variational inference framework OCEAN to perform online task inference for compositional tasks. OCEAN models global and local context variables in a joint latent space, where the global variables represent a mixture of subtasks required for the task, while the local variables capture the transitions between the subtasks. Our framework supports flexible latent distributions based on prior knowledge of the task structure and can be trained in an unsupervised manner. Experimental results show that OCEAN provides more effective task inference with sequential context adaptation and thus leads to a performance boost on complex, multi-stage tasks.« less
  5. Scientific data analysis pipelines face scalability bottlenecks when processing massive datasets that consist of millions of small files. Such datasets commonly arise in domains as diverse as detecting supernovae and post-processing computational fluid dynamics simulations. Furthermore, applications often use inference frameworks such as TensorFlow and PyTorch whose naive I/O methods exacerbate I/O bottlenecks. One solution is to use scientific file formats, such as HDF5 and FITS, to organize small arrays in one big file. However, storing everything in one file does not fully leverage the heterogeneous data storage capabilities of modern clusters. This paper presents Henosis, a system that interceptsmore »data accesses inside the HDF5 library and transparently redirects I/O to the in-memory Redis object store or the disk-based TileDB array store. During this process, Henosis consolidates small arrays into bigger chunks and intelligently places them in data stores. A critical research aspect of Henosis is that it formulates object consolidation and data placement as a single optimization problem. Henosis carefully constructs a graph to capture the I/O activity of a workload and produces an initial solution to the optimization problem using graph partitioning. Henosis then refines the solution using a hill-climbing algorithm which migrates arrays between data stores to minimize I/O cost. The evaluation on two real scientific data analysis pipelines shows that consolidation with Henosis makes I/O 300× faster than directly reading small arrays from TileDB and 3.5× faster than workload-oblivious consolidation methods. Moreover, jointly optimizing consolidation and placement in Henosis makes I/O 1.7× faster than strategies that perform consolidation and placement independently.« less