Early Warning and Response System (EWRS) systems aim to: Identify the causes of a conflict, anticipate their outbreak, and mitigate their impact. The EWRS are mechanisms for preventing and addressing conflicts that focus on the systematic collection, processing and analysis of information (quantitative or qualitative) about conflict situations for the purpose of warning decision-makers so that they can take measures or implement actions that will avoid the emergence or escalation of conflict. Modern technology enables us to predict the potential conflict events the world shall see in the future, it can also single out specific regions and countries for special attention. This allow organizations to perform early warning risk assessments for better planning and to target non-conflict interventions potentially limiting the break out or spread of conflict.
DARPA launched in 2017, the World Modelers program, with a aim to develop technology that will enable analysts to rapidly build models to analyze questions relevant to national and global security. Questions for analysis will typically be framed at subnational scales and look one to five years into the future, although the factors that influence outcomes of interest might operate on larger spatial and temporal scales.
Scientists and analysts, whether they are from the intelligence community or non-governmental organisations, typically convene a panel of domain experts across a variety of different fields, such as agriculture, regional security, and economics and markets, to gather the appropriate data sets and models and run them, Elliot said.
“Two years later after starting this project they will release a report explaining exactly what happened, what caused the famine, and what interventions could prevent it. But of course, by that point it has been two years and the issue is already well past,” he said. The hope is that the analysis will help inform decision makers when the next event arises but because these are relatively rare events, it could be a decade before the next one happens and by then a report could be long forgotten or no longer valid.
The purpose of the World Modelers program is to develop technology that will enable analysts to rapidly build models to analyze questions relevant to national and global security. Questions for analysis will typically be framed at subnational scales and look one to five years into the future, although the factors that influence outcomes of interest might operate on larger spatial and temporal scales. The predictive type models will enable different scenarios to be analysed as well as analysis of various intervention strategies.
World Modelers aims to enable a single analyst to conduct in about a month work that had previously taken two years to do. That analyst would identify complex sets and interventions that could solve or reduce the impact of a crisis while it is emerging.
The goal of DARPA’s World Modelers effort is to reduce the amount of time it takes to examine an issue, analyse it, come up with solutions and responses, and then write a report, Joshua Elliot, programme director for DARPA’s World Modelers, told Jane’s .
World Modelers analyses are intended to be timely enough to recommend specific actions that could avert crises. The program seeks to develop technologies that will provide clearly parameterized, quantitative projections within weeks or even hours of processing, compared to the months or years it takes today to understand considerably simpler systems.
World Modelers technologies will be applied to increasingly varied use cases as they mature through the phases of the program. Questions for analysis will typically be framed at subnational scales and look one to five years into the future, although the factors that influence outcomes of interest might operate on larger spatial and temporal scales. This subnational focus reflects the changing nature of conflict and security, which, increasingly, plays out in cities and districts. The first use case of World Modelers is food insecurity resulting from the interactions of multiple factors, including climate, water availability, soil viability, market instability, and physical security.
The goal of the World Modelers Program is to construct technology that allows the rapid building of models to address questions pertaining to both national and global security. The program will focus on food insecurity as an initial test case in phase 1 with the aim to build models that can predict future food insecurity in specific geographic locations
Food insecurity is a growing problem in many parts of the world where societies struggle to feed growing populations. Food insecurity has both humanitarian challenges and regional security ramifications, as food shortages can result in migration/displacement and conflict between peoples. Food insecurity will be the initial use case for the World Modelers program, and World Modelers technologies will enable researchers to efficiently and rigorously perform analytic tasks such as “Analyze food insecurity in each district of South Sudan two years into the future.” World Modelers technologies will be applied to additional (and increasingly varied) use cases as they mature through a sequence of program phases.
Large organizations have spent years perfecting analytical methods that do some of the above. World Modelers technology is expected to build models to solve this problem – and many others like it – in a month.
World Modelers Program
On March 27, the Defense Advanced Research Projects Agency (DARPA) Information Innovation Office (I20) released the Broad Agency Announcement (BAA) for the World Modelers Program, led by program manager Dr. Paul Cohen.
DARPA is soliciting innovative research proposals in the area of causal modeling, forecasting, and analysis techniques. Proposed research should investigate innovative approaches that enable revolutionary advances in science, mathematics, or technology. Specifically excluded is research that primarily results in evolutionary improvements to the existing state of practice.
The World Modelers program aims to develop technology that integrates qualitative causal analyses with quantitative models and relevant data to provide a comprehensive understanding of complicated, dynamic national security questions. The goal is to develop approaches that can accommodate and integrate dozens of contributing models connected by thousands of pathways—orders of magnitude beyond what is possible today.
Much of the work is based on advanced domain agnostic machine reading-type technologies, Elliot explained. “A lot of the work we expect to happen under the programme is looking at how we can automatically extract elaborative causal models with contextual information … arbitrary national and global security situations, such as food security or unrest,” he said. “How can we translate those causal models into quantitative work flows by bringing in relevant data … by bringing in relevant quantitative models that are domain specific, [and] by coupling these modes together.”
A lot of the data that World Modelers will use will come from satellite imagery. The programme’s spatial scale will cover regional all the way down to the ability to identify security situations happening at the local level. World Modelers will also rely on historical and survey data to provide a highly granular representation of an affected population and area.
DARPA provides a “Food Shortage Forecast Example,” For any subnational location (e.g., Southern Sudan) generate food shortage scenarios two years out. Consider a comprehensive set of causes. Integrate current and historical data and human expert analysis. Model the scenarios quantitatively and probabilistically. Make forecasts, explain sources of uncertainty, and identify and explain specific potential solutions. Update models, forecasts and probabilities on receipt of new data.
There are five Technical Areas (TAs) of focus:
TA1: Build Qualitative Models from Online Sources – construct “qualitative, causal analysis graphs semi-automatically.” This builds on Big Mechanism reading for causal fragments, grounding in ontologies, assembly algorithms.
TA2: Workflow Compiler (for Integration of Quantitative Models) – compile open-source quantitative models pertaining to the proposers’ selected use case and integrate models into a graphical analysis. This involves scientific workflow technology (e.g., WINGS), grid computing, distributed simulation, CwC collaborative dialog capability.
TA3: Parameterize Models – address the challenge of automating techniques to “convert data into values of parameters of quantitative models, aligned with specific analyses.” Big Mechanism table reading, remote sensing feeds, crowdsourcing and polling.
TA4: From Scenarios to Actions – define ways in which “specifying scenarios and analyzing actions will be shared between machines and humans.” Big Mechanism tech for finding “pressure points” in causal networks, AI planning tech for multi-step plans TA5: Uncertainty Reports – provide analysts and clients “reasons to believe (or disbelieve) analyses.”
SRI International Wins Contract for DARPA Modeling Program Support
SRI International has won a $7.2 million contract from the U.S. Army to support a Defense Advanced Research Project Agency program that aims to develop modeling technology for analysts to examine queries related to national and global security.
The company will support DARPA’s World Modelers Program and is scheduled to finish work under the cost-plus-fixed-fee contract by Dec. 9, 2021, the Defense Department said. DoD noted the program seeks research proposals in the causal modeling, forecasting and analysis areas.
The Army Contracting Command received seven bids for the contract and the service branch obligated $1.4 million in fiscal 2017 research and development funds at the time of award.
ISI Researchers to “Model the World” with New DARPA Award
USC Viterbi’s Information Sciences Institute researchers are leveraging artificial intelligence to create an ambitious new computer modeling and predictive analysis tool thanks to a new $13 million Defense Research Projects Award (DARPA) award.
Principal investigator Yolanda Gil, an ISI research director and USC Viterbi computer science professor, and a team of scientists and researchers are developing a tool to better understand the evolving relationship between humans and nature and its impact on the planet.
The four-year project, called MINT for Model INTegration, is part of DARPA’s World Modelers program.
Using artificial intelligence (AI) techniques, the system will combine existing models from various disciplines-including hydrology, agriculture, economics and social sciences-and run simulations to predict the consequences of actions or policies in specific geographic regions.
For instance, what could be the social and economic impacts of dam construction on fisheries and crop production over time?
The project’s co-principal investigators are ISI researchers and USC Viterbi faculty members Ewa Deelman, a research professor of computer science; Rafael Ferreira da Silva, a computer science research assistant professor; and Craig Knoblock, a computer science research professor.
The science of prediction
While big data collection is becoming an increasingly automated process, complex world models are currently still pieced together by hand. This is a particularly challenging task, since many problems require the integration of models from different fields with diverse modeling approaches. In fact, analyzing just one scenario in this way can take up to two years. Using automation and artificial intelligence, Gil and her team hope to bring that down to a couple of days. The team includes researchers from Virginia Tech, the Pennsylvania State University, the University of Minnesota and the University of Colorado.
Building on group’s extensive prior work on model integration both from disciplinary and computational perspectives, the researchers have proposed a comprehensive approach, which includes cross-disciplinary modeling, model integration, workflow composition and execution, data extraction and data integration.
“We use AI to process and describe the data and information in a very precise way,” said Gil, who leads ISI’s Interactive Knowledge Capture Group.
“We develop ontologies to describe specific terms and knowledge graphs to help us relate those terms to others. Then, we apply machine learning algorithms to refine the data and workflow tools to track the data flow through the software. The output is the prediction, for example, water availability in a specific region.”
Modeling experts will provide test scenarios, as well as data sources and models, including data extracted from remote Earth observation. The team plans to open-source the MINT software to create a community of users that will continue to use and develop the software over time.
“Ultimately, by developing this sophisticated modeling environment, MINT aims to significantly reduce the time needed to develop new integrated models,” said Gil. “We hope to generate a better understanding of how human and natural processes impact one another—that’s the holy grail.”