WASHINGTON — Taking seemingly unrelated events and seeing a pattern is a key to successful sleuthing after the fact. The Pentagon hopes to come up with a way to collect seemingly random facts and determine their collective conclusion — before that conclusion occurs.
To achieve that foresight, DARPA is launching an effort to use artificial intelligence (AI) to help uncover complex and superficially unrelated events found throughout information sources, then collate them for the user to take proactive measures. In short, DARPA hopes to have clear knowledge to change an outcome before that outcome can unfold.
“Rapid comprehension of world events is critical to informing national security efforts,” DARPA said in a release.
DARPA stands for the Defense Advanced Research Projects Agency. It is the Pentagon’s research, innovation, and “no-bad-ideas” wing.
““The process of uncovering relevant connections across mountains of information and the static elements that they underlie requires temporal information and event patterns, which can be difficult to capture at scale with currently available tools and systems,” Boyan Onyshkevych, a program manager in DARPA’s Information Innovation Office, said in a news release.
The new program is called the Knowledge-directed Artificial Intelligence Reasoning Over Schemas (KAIROS) program.
“KAIROS seeks to create a schema-based AI capability to enable contextual and temporal reasoning about complex real-world events in order to generate actionable understanding of these events and predict how they will unfold,” DARPA said in its news release. “The program aims to develop a semi-automated system capable of identifying and drawing correlations between seemingly unrelated events or data, helping to inform or create broad narratives about the world around us.”
The first step will be to have AI create schemas — knowledge that humans mentally log to give understanding to events by organizing them into commonly occurring narrative structures. The theory is that AI could create schemas from “large volumes of data by detecting, classifying and clustering sub-events based on linguistic inference and common sense reasoning” that humans alone cannot do, DARPA said.
The second phase will pivot on AI applying the “library of schemas” created during the first step across the media and language spectrums to detect, snare and then process “events and entities, as well as relationships among them to help construct and extend a knowledge base,” DARPA.
DARPA will hold a Proposers Day notice on the new program on January 9, calling for ideas and to provide more information about KAIROS and answer questions from potential proposers.