MIT scientists have developed a new type of "deep learning" network that can help robots measure the quality of their interactions with children with autism spectrum conditions by using specific data for each child.
Autism Spectrum Disorder is a condition of brain development that affects how a person perceives and socializes with others, resulting in problems in social interaction and communication. The term "spectrum" in Autism Spectrum Disorder refers to the wide range of symptoms and severity.
Armed with personalized "deep learning", the kid-friendly robot NAO can use data to estimate the engagement and interest of every autistic child. The result is a unique study for 35 autistic children.
The new development can make your life easier.
"The long-term goal is not to create robots that replace human therapists, but to supplement them with key information that therapists can use to personalize therapy content as well as more interesting and naturalistic interactions between robots and children to do with autism, "said Oggi Rudovic, the lead author of the study.
The perception of the robot, the answers of the children agreed with the assessments of human experts, according to the scientists a correlation of 60 percent.
"The Challenge of Creating Machine Learning and Artificial Intelligence [artificial intelligence] tha t is particularly annoying in autism because the standard AI methods require a lot of data that is similar for each class learned in autism, where heterogeneity The normal AI approaches fail, "said Rosalind Picard, co-author of the study.
In robot-assisted therapy, a human therapist displays flash cards of various faces expressing different emotions, such as happiness, sadness, anxiety, and a programmed robot or NAO that shows the child the same feelings.
The therapist observes how the child deals with the robot and receives feedback on how the lesson should be continued.
NAO is a two-foot robot reminiscent of an armored superhero or droid that transmits different emotions through changes in eye color, limb movement and tone.
"Therapists say it can take a few seconds to challenge them, and robots attract the child's attention," said Rudovic.
"People change their expressions in many different ways, but the robots always do it in the same way, and that's less frustrating for the child, because the child learns in a very structured way, as the expressions are shown," he said.
The system created by these researchers includes not only personalization, but also deep learning conducted by AI.
Deep learning has become automatic speech and object recognition programs, making it a good fit for a problem, such as the various facial, body, and vocal traits that help to understand a more abstract concept, such as a child's engagement.
"Deep learning allows the robot to extract the most important information directly from this data without people having to manually create these features," says Rudovic.
For the therapy robots a personalized framework was created, which can learn from the collected data individual child.
(This story was not edited by devdiscourse staff and is automatically generated from a shared feed.)