The machine learning system is being developed as part of a $6million contract for the Competency Aware Machine Learning programme (CAML) from the Defense Advanced Research Projects Agency (DARPA). Systems will be able to communicate the abilities they have learned, the conditions under which the abilities were learned, the strategies they recommend and the situations for which those strategies can be used.
The system will learn from a video game like process. The system will explain itself by recording the conditions and strategies it used to come up with successful outcomes.
Principal investigator for CAML at Raytheon BBN Technologies, Ilana Heintz, said: “The CAML system turns tools into partners. It will understand the conditions where it makes decisions and communicate the reasons for those decisions.”
“People need to understand an autonomous system’s skills and limitations to trust it with critical decisions.”
Once the system has developed these skills, the team will apply it to a simulated search and rescue mission. Users will create the conditions surrounding the mission, while the system will make recommendations and give users information about its competence in those particular conditions.
For example, the system might say, ‘In the rain, at night, I can distinguish between a person and an inanimate object with 90 percent accuracy, and I have done this over 1,000 times’. Raytheon have explained that the goal is to create trust between AI systems and human operators through user-friendly communication.
Defence Online is the official media partner of DPRTE 2020. To book your exhibition or sponsorship package please contact the team today on 0845 270 7066 or exhibitions@dprte.co.uk
Limited offer: Delegate rates are being held at 2019 prices until 31 January. To book for just £95 click here
If you would like to join our community and read more articles like this then please click here.
ai DARPA Information machine intelligence Raytheon Raytheon BBN Technologies Search and Rescue