Back to news

February 18, 2019 | International, C4ISR

DARPA: Teaching AI Systems to Adapt to Dynamic Environments

Current AI systems excel at tasks defined by rigid rules – such as mastering the board games Go and chess with proficiency surpassing world-class human players. However, AI systems aren't very good at adapting to constantly changing conditions commonly faced by troops in the real world – from reacting to an adversary's surprise actions, to fluctuating weather, to operating in unfamiliar terrain. For AI systems to effectively partner with humans across a spectrum of military applications, intelligent machines need to graduate from closed-world problem solving within confined boundaries to open-world challenges characterized by fluid and novel situations.

To attempt this leap, DARPA today announced the Science of Artificial Intelligence and Learning for Open-world Novelty (SAIL-ON) program. SAIL-ON intends to research and develop the underlying scientific principles and general engineering techniques and algorithms needed to create AI systems that act appropriately and effectively in novel situations that occur in open worlds. The program's goals are to develop scientific principles to quantify and characterize novelty in open-world domains, create AI systems that react to novelty in those domains, and to demonstrate and evaluate these systems in a selected DoD domain. A Proposers Day for interested proposers is scheduled for March 5, 2019, in Arlington, Virginia: https://go.usa.gov/xEUWh

“Imagine if the rules for chess were changed mid-game,” said Ted Senator, program manager in DARPA's Defense Sciences Office. “How would an AI system know if the board had become larger, or if the object of the game was no longer to checkmate your opponent's king but to capture all his pawns? Or what if rooks could now move like bishops? Would the AI be able to figure out what had changed and be able to adapt to it?”

Existing AI systems become ineffective and are unable to adapt when something significant and unexpected occurs. Unlike people, who recognize new experiences and adjust their behavior accordingly, machines continue to apply outmoded techniques until they are retrained.

Given enough data, machines are able to do statistical reasoning well, such as classifying images for face-recognition, Senator said. Another example is DARPA's AI push in self-driving cars in the early 2000s, which led to the current revolution in autonomous vehicles. Thanks to massive amounts of data that include rare-event experiences collected from tens of millions of autonomous miles, self-driving technology is coming into its own. But the available data is specific to generally well-defined environments with known rules of the road.

“It wouldn't be practical to try to generate a similar data set of millions of self-driving miles for military ground systems that travel off-road, in hostile environments and constantly face novel conditions with high stakes, let alone for autonomous military systems operating in the air and on sea,” Senator said.

If successful, SAIL-ON would teach an AI system how to learn and react appropriately without needing to be retrained on a large data set. The program seeks to lay the technical foundation that would empower machines, regardless of the domain, to go through the military OODA loop process themselves – observe the situation, orient to what they observe, decide the best course of action, and then act.

“The first thing an AI system has to do is recognize the world has changed. The second thing it needs to do is characterize how the world changed. The third thing it needs to do is adapt its response appropriately,” Senator said. “The fourth thing, once it learns to adapt, is for it to update its model of the world.”

SAIL-ON will require performers and teams to characterize and quantify types and degrees of novelty in open worlds, to construct software that generates novel situations at distinct levels of a novelty hierarchy in selected domains, and to develop algorithms and systems that are capable of identifying and responding to novelty in multiple open-world domains.

SAIL-ON seeks expertise in multiple subfields of AI, including machine learning, plan recognition, knowledge representation, anomaly detection, fault diagnosis and recovery, probabilistic programming, and others. A Broad Agency Announcement (BAA) solicitation is expected to be posted in the near future and will be available on DARPA's FedBizOpps page: http://go.usa.gov/Dom

https://www.darpa.mil/news-events/2019-02-14

On the same subject

  • General Dynamics unit wins contract to build new light tank for infantry

    June 30, 2022 | International, Land

    General Dynamics unit wins contract to build new light tank for infantry

    General Dynamics Land Systems beats out BAE Systems in a competition to build and field Mobile Protected Firepower systems for the Infantry Brigade Combat Team.

  • Defense Industry’s Covid Closings Decline, Pentagon Agency Says

    July 8, 2020 | International, Aerospace, Naval, Land, C4ISR, Security

    Defense Industry’s Covid Closings Decline, Pentagon Agency Says

    By Anthony Capaccio The defense industry has made major strides reducing the impact of Covid-19 on operations, decreasing total closings of facilities to six on Monday from a high of 148 in mid-April, according to the Pentagon agency that oversees contracts. “We're seeing a significantly smaller fraction of the industrial base impacted on a daily basis” as contractors have become “better at restoring operational capability after potential exposures” to the coronavirus, Army Lieutenant General David Bassett, director of the Defense Contract Management Agency, said in an interview. “We've gone from having a substantial fraction of the industrial base impacted to today,” where it's “just a handful.” In total, 279 defense contracting locations were forced to shut down an average of 20 days since April because of the pandemic. In addition, 149 locations currently have reduced operations because of the virus, according to the agency, which tracks 10,509 locations of major defense contractors and their subcontractors. “These closures have generally been short-term in order to clean facilities” or to “reduce the potential exposure of employees,” according to agency spokesman Matthew Montgomery. Ellen Lord, the Defense Department's acquisitions chief, has warned that pandemic disruptions are expected to result in defense industry claims for reimbursement of more than $10 billion under the Cares Act, which provides economic aid including reimbursing contractors for payments to employees affected by disruptions such as plant closings. She has said a single contractor, which she didn't name, is estimated to have at least $1.5 billion in potential claims. Bassett said the decline in plant closings reflects that companies “have really got a plan in place so that they know what they have to do when they find people who have been exposed, how they have to handle the plant and then what they can do to get back up quickly and safely.” Bassett assumed command of the contract agency on June 3 after a career that included positions as the Army's top program manager for command-and-control networks and for ground-combat vehicles. “As we watch right now and cases are beginning to rise in certain areas of the country, I've asked all of our teams to really think about what we can do right now to make sure if we do end up in a shutdown we can avoid impacts to the industrial base and our deliveries,” he said. https://www.bloomberg.com/news/articles/2020-07-07/defense-industry-s-covid-closings-decline-pentagon-agency-says

  • Germany OKs buying howitzers to help replace arms rushed to Ukraine

    March 29, 2023 | International, Land

    Germany OKs buying howitzers to help replace arms rushed to Ukraine

    The German parliament on Wednesday approved defence ministry plans to buy up to 28 self-propelled howitzers to replace weapons rushed to Ukraine out of army stocks last year, lawmakers familiar with the matter told Reuters.

All news