Quantcast
Channel: Argonne Today
Viewing all articles
Browse latest Browse all 2888

Artificial Intelligence and High Performance Computing Journal Club seminar series

$
0
0

The AI/HPC Journal Club will hold a seminar on Friday, Oct. 18, 2019, at 10 a.m. in Building 241, Room D172. Join us for a presentation by Jonathan Ozik (DIS).

“Harnessing Complex Systems with Agent-based Modeling, Machine Learning and High-performance Computing”

Abstract

In this presentation I will review efforts by our research group to formally facilitate the intersection of agent-based modeling, machine learning methods, and high-performance computing, three areas of continuing general interest and growth, to tackle the intricacies of complex systems modeling. I will provide a brief overview of agent-based modeling and discuss our widely used, free and open source Repast Suite of agent-based modeling toolkits (https://repast.github.io). I will then describe how our Extreme-scale Model Exploration with Swift (EMEWS) framework (https://emews.github.io) leverages advances in machine learning algorithms to enable large-scale model exploration of computational models, including agent-based models, on high-performance computing resources. I will demonstrate applications of our approach across scientific domains where the three pillars of agent-based modeling, machine learning and high-performance computing provide the analytical platform for in silico experiments at the scales needed for deepening our understanding of important complex systems phenomena. Finally, I will describe future directions and our overarching goal of improving interoperability, scalability, transparency and reproducibility in complex systems modeling.

For additional information, contact Sijia Dong.

Join the seminar on BlueJeans.

Add to Your Calendar


Viewing all articles
Browse latest Browse all 2888

Trending Articles