• Corpus ID: 238583327

DeepABM: Scalable, efficient and differentiable agent-based simulations via graph neural networks

  title={DeepABM: Scalable, efficient and differentiable agent-based simulations via graph neural networks},
  author={Ayush Chopra and Esma Senturk Gel and Jayakumar Subramanian and Balaji Krishnamurthy and Santiago Romero-Brufau and Kalyan Sunder Pasupathy and Thomas C. Kingsley and Ramesh Raskar},
We introduce DeepABM, a framework for agent-based modeling that leverages geometric message passing of graph neural networks for simulating action and interactions over large agent populations. Using DeepABM allows scaling simulations to large agent populations in real-time and running them efficiently on GPU architectures. To demonstrate the effectiveness of DeepABM, we build DeepABM-COVID simulator to provide support for various non-pharmaceutical interventions (quarantine, exposure… 

Figures and Tables from this paper

Application of Deep Learning Techniques in Diagnosis of Covid-19 (Coronavirus): A Systematic Review

DL-based Covid-19 detection systems are the key focus of this review article, evaluating causal agents, pathophysiology, immunological reactions, and epidemiological illness.

Physics vs. Learned Priors: Rethinking Camera and Algorithm Design for Task-Specific Imaging

This paper presents a framework to understand the building blocks of this nascent field of end-to-end design of camera hardware and algorithms, and shows how methods that exploit both physics and data have become prevalent in imaging and computer vision.



OpenABM-Covid19—An agent-based model for non-pharmaceutical interventions against COVID-19 including contact tracing

OpenABM-Covid19 is presented: an agent-based simulation of the epidemic including detailed age-stratification and realistic social networks and its Python interface has allowed scientists and policymakers to simulate dynamic packages of interventions and help compare options to suppress the COVID-19 epidemic.

Hierarchical Message-Passing Graph Neural Networks

Empirical experiments exhibit that HC-GNN can outperform state-of-the-art GNN models in network analysis tasks, including node classification, link prediction, and community detection, and the model analysis further demonstrates HC- GNN’s robustness facing graph sparsity and the flexibility in incorporating different GNN encoders.

Graph Attention Networks

We present graph attention networks (GATs), novel neural network architectures that operate on graph-structured data, leveraging masked self-attentional layers to address the shortcomings of prior

Mesa: An Agent-Based Modeling Framework

Mesa is a new open-source, Apache 2.0 licensed package that allows users to quickly create agent-based models using built-in core components or customized implementations; visualize them using a browser-based interface; and analyze their results using Python's data analysis tools.

PyTorch: An Imperative Style, High-Performance Deep Learning Library

This paper details the principles that drove the implementation of PyTorch and how they are reflected in its architecture, and explains how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance.

Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing

A mathematical model for infectiousness was developed to estimate the basic reproductive number R0 and to quantify the contribution of different transmission routes and the requirements for successful contact tracing, and the combination of two key parameters needed to reduce R0 to less than 1 was determined.

Authors’ response: Estimating the generation interval for COVID-19 based on symptom onset data

Estimating generation and serial interval distributions from outbreak data requires careful investigation of the underlying transmission network, which is essential for correctly estimating these quantities.

Modeling the combined effect of digital exposure notification and non-pharmaceutical interventions on the COVID-19 epidemic in Washington state

In a model in which 15% of the population participated, it is found that digital exposure notification systems could reduce infections and deaths by approximately 8% and 6%, effectively complementing traditional contact tracing.

Markov Games as a Framework for Multi-Agent Reinforcement Learning

Semi-Supervised Classification with Graph Convolutional Networks

A scalable approach for semi-supervised learning on graph-structured data that is based on an efficient variant of convolutional neural networks which operate directly on graphs which outperforms related methods by a significant margin.