Raytheon subsidiary BBN Technologies is developing a new and advanced neural network under the US Defense Research Project Agency’s (DARPA) Explainable Artificial Intelligence (XAI) programme.
The new Explainable Question Answering System (EQUAS) is a first-of-its-kind system that has been designed to enable artificial intelligence (AI) to explain itself.
The XAI programme focuses on developing a range of new machine-learning techniques that can provide more explainable models while ensuring an increased level of performance.
In addition, it enables human users to understand, appropriately trust and effectively manage the future generation of artificially intelligent partners.
Raytheon BBN Technologies lead scientist and EQUAS principal investigator Bill Ferguson said: “Our goal is to give the user enough information about how the machine’s answer was derived and show that the system considered relevant information so users feel comfortable acting on the system’s recommendation.”
Currently in its early phases of development, the EQUAS technology will show users which data was most significant and important in the AI decision-making process.
The users can use a graphical interface to evaluate the recommendations provided by the BBN system and know why one answer is chosen over another.
Upgrades on the EQUAS system will enable the technology to monitor itself and share factors that limit its ability to make reliable recommendations.
The self-monitoring ability of the Raytheon solution will enable developers to refine the AI systems, thereby enabling them to provide additional data or change how data is processed.