Improving Effectiveness of Simulation-Based Inference in the Massively Parallel Regime

Simulation-based Inference (SBI) is a widely used set of algorithms to learn the parameters of complex scientific simulation models. While primarily run on CPUs in High-Performance Compute clusters, these algorithms have been shown to scale in performance when developed to be run on massively parallel architectures such as GPUs. While parallelizing existing SBI algorithms provides us with performance gains, this might not be the most efficient way to utilize the achieved parallelism. This work proposes a new parallelism-aware adaptation of an existing SBI method, namely approximate Bayesian computation with Sequential Monte Carlo(ABC-SMC). This new adaptation is designed to utilize the parallelism not only for performance gain, but also toward qualitative benefits in the learnt parameters. The key idea is to replace the notion of a single ‘step-size’hyperparameter, which governs how the state space of parameters is explored during learning, with step-sizes sampled from a tuned Beta distribution. This allows this new ABC-SMC algorithm to more efficiently explore the state-space of the parameters being learned. We test the effectiveness of the proposed algorithm to learn parameters for an epidemiology model running on a Tesla T4 GPU. Compared to the parallelized state-of-the-art SBI algorithm, we get similar quality results in <inline-formula><tex-math notation="LaTeX">$\sim 100 \times$</tex-math></inline-formula> fewer simulations and observe <inline-formula><tex-math notation="LaTeX">$\sim 80 \times$</tex-math></inline-formula> lower run-to-run variance across 10 independent trials. IEEE

Hardware-accelerated Simulation-based Inference of Stochastic Epidemiology Models for COVID-19

Epidemiology models are central to understanding and controlling large-scale pandemics. Several epidemiology models require simulation-based inference such as Approximate Bayesian Computation (ABC) to fit their parameters to observations. ABC inference is highly amenable to efficient hardware acceleration. In this work, we develop parallel ABC inference of a stochastic epidemiology model for COVID-19. The statistical inference framework is implemented and compared on Intel's Xeon CPU, NVIDIA's Tesla V100 GPU, Google's V2 Tensor Processing Unit (TPU), and the Graphcore's Mk1 Intelligence Processing Unit (IPU), and the results are discussed in the context of their computational architectures. Results show that TPUs are 3×, GPUs are 4×, and IPUs are 30× faster than Xeon CPUs. Extensive performance analysis indicates that the difference between IPU and GPU can be attributed to higher communication bandwidth, closeness of memory to compute, and higher compute power in the IPU. The proposed framework scales across 16 IPUs, with scaling overhead not exceeding 8% for the experiments performed. We present an example of our framework in practice, performing inference on the epidemiology model across three countries and giving a brief overview of the results. © 2022 Association for Computing Machinery.

A hybrid stochastic model and its Bayesian identification for infectious disease screening in a university campus with application to massive COVID-19 screening at the University of Liège.

Amid the COVID-19 pandemic, universities are implementing various prevention and mitigation measures. Identifying and isolating infectious individuals by using screening testing is one such a measure that can contribute to reducing spread. Here, we propose a hybrid stochastic model for infectious disease transmission in a university campus with screening testing and its surrounding community. Based on a compartmental modeling strategy, this hybrid stochastic model represents the evolution of the infectious disease and its transmission using continuous-time stochastic dynamics, and it represents the screening testing as discrete stochastic events. We also develop, in a Bayesian framework, the identification of parameters of this hybrid stochastic model, including transmission rates. These parameters were identified from the screening test data for the university population and observed incidence counts for the surrounding community. We implement the exploration of the Bayesian posterior using a machine-learning simulation-based inference approach. The proposed methodology was applied in a retrospective modeling study of a massive COVID-19 screening conducted at the University of Liège in Fall 2020. The emphasis of the paper is on the development of the hybrid stochastic model to assess the impact of screening testing as a measure to reduce spread. The hybrid stochastic model allows various factors to be represented and examined, such as interplay with the surrounding community, variability of the transmission dynamics, the rate of participation in the screening testing, the test sensitivity, the test frequency, the diagnosis delay, and compliance with isolation. The application in the retrospective modeling study suggests that a high rate of participation and a high test frequency are important factors to reduce spread.