17 Apr 2024

Enabling Understanding of Artificial Intelligence (AI) Agent Wargaming Decisions through Visualizations

Abstract: The process to develop options for military planning course of action (COA) development and analysis relies on human subject matter expertise. Analyzing COAs requires examining several factors and understanding complex interactions and dependencies associated with actions, reactions, proposed counteractions, and multiple reasonable outcomes. In Fiscal Year 2021, the Institute for Systems Engineering Research team completed efforts resulting in a wargaming maritime framework capable of training an artificial intelligence (AI) agent with deep reinforcement learning (DRL) techniques within a maritime scenario where the AI agent credibly competes against blue agents in gameplay. However, a limitation of using DRL for agent training relates to the transparency of how the AI agent makes decisions. If leaders were to rely on AI agents for COA development or analysis, they would want to understand those decisions. In or-der to support increased understanding, researchers engaged with stakeholders to determine visualization requirements and developed initial prototypes for stakeholder feedback in order to support increased understanding of AI-generated decisions and recommendations. This report describes the prototype visualizations developed to support the use case of a mission planner and an AI agent trainer. The prototypes include training results charts, heat map visualizations of agent paths, weight matrix visualizations, and ablation testing graphs.

2 Dec 2022

In Situ and Time

Abstract: Large-scale HPC simulations with their inherent I/O bottleneck have made in situ visualization an essential approach for data analysis, although the idea of in situ visualization dates back to the era of coprocessing in the 1990s. In situ coupling of analysis and visualization to a live simulation circumvents writing raw data to disk for post-mortem analysis -- an approach that is already inefficient for today's very large simulation codes. Instead, with in situ visualization, data abstracts are generated that provide a much higher level of expressiveness per byte. Therefore, more details can be computed and stored for later analysis, providing more insight than traditional methods. This workshop encouraged talks on methods and workflows that have been used for large-scale parallel visualization, with a particular focus on the in situ case.

26 Jan 2022

In Situ Visualization with Temporal Caching

Abstract: In situ visualization is a technique in which plots and other visual analyses are performed in tandem with numerical simulation processes in order to better utilize HPC machine resources. Especially with unattended exploratory engineering simulation analyses, events may occur during the run, which justify supplemental processing. Sometimes though, when the events do occur, the phenomena of interest includes the physics that precipitated the events and this may be the key insight into understanding the phenomena that is being simulated. In situ temporal caching is the temporary storing of produced data in memory for possible later analysis including time varying visualization. The later analysis and visualization still occurs during the simulation run but not until after the significant events have been detected. In this article, we demonstrate how temporal caching can be used with in-line in situ visualization to reduce simulation run-time while still capturing essential simulation results.

30 Sep 2021

Alternative Analysis for Construction Progress Data Spatial Visualization

Abstract: The U.S. Army Corps of Engineers (USACE) construction projects have multiple stakeholders that collaborate with project delivery team members during the execution of these projects. Many of these stakeholders are located across the U.S., which makes virtual interactions a common communication method for these teams. These interactions often lack spatial visualization, which can add complications to the progress reports provided and how the information is received/interpreted. The visualization of project progress and documents would be invaluable to the stakeholders on critical projects constructed by the USACE. This research was conducted to determine alternatives for migrating Resident Management System (RMS) data into a portal web viewer. This report provides proposed solutions to creating these links in efforts to better harmonize data management and improve project presentation.

8 Jun 2021

Summary of the SciTech 2020 Technical Panel on In Situ/In Transit Computational Environments for Visualization and Data Analysis

Link: http://dx.doi.org/10.21079/11681/40887This paper was originally presented at the American Institute of Aeronautics and Astronautics (AIAA) ScitTech 2020 Technical Panel and published online 4 January 2021. Funding by USACE ERDC under Army Direct funding.Report Number: ERDC/ITL MP-21-10Title: Summary of the SciTech 2020 Technical Panel on In

17 Jun 2020

PUBLICATION NOTICE: Water Quality Visualization Tools: A Python Application (1/A)

Abstract: On May 4, 2016, US District Court Judge Simon ordered the US Army Corps of Engineers and two other Action Agencies to produce a comprehensive Environmental Impacts Statement (EIS) by March 26, 2021. To do this, the Columbia River Systems Operation (CRSO) EIS will evaluate and compare a range of alternatives to offset or minimize any remaining unavoidable impacts. Due to the unique large system model approach, there is a need to quickly develop and analyze water quality model results. Therefore, there was a need for several visualization tools to assist the CRSO EIS team in promptly analyzing the results and creating publication-ready graphics. To create the most accessible desktop application for the CRSO EIS team, the Python programming language was used to quickly create three visualization tools. These three tools are only useful for relatively small data sets. If the team wishes to expand the functionality for larger data sets, it is recommended that model execution and analysis be moved to the supercomputers.