[ad_1]
MIT scientists have developed a new kind of in-depth learning network that can help robots evaluate the quality of their interactions with children with autism spectrum disorders by using data specific to each child . brain development that influences the way a person perceives and socializes with others, causing problems of social interaction and communication. The term "spectrum" in autism spectrum disorder refers to the wide range of symptoms and severity.
Armed with a "deep learning" custom, the NAO child-friendly robot can easily gauge the commitment and interest of each autistic child. The new development can make their life easier.
"The long-term goal is not to create robots that will replace human therapists, but to reinforce them with key information that therapists can use to personalize the content of the therapy and also make more engaging and naturalistic interactions between robots and autistic children, "said Oggi Rudovic, the first author of the study
. children's responses were in agreement with human expert assessments, with a correlation of 60%, according to scientists.
"The Challenge of Creating Machine Learning and AI [artificial intelligence] The work on autism is particularly vexing because the usual AI methods require a lot of similar data for Each category learned In autism, where heterogeneity reigns, normal approaches to AI fail, says Rosalind Picard, co-author of the study
Assisted Therapy by robot involves a human therapist showing flash cards of different faces expressing different emotions, such as happiness, sadness, fear and a programmed robot or NAO, which shows the same emotions to the child.
The therapist observes how the child interacts with the robot and gets feedback on how the lesson should be pushed further
NAO a two-foot robot resembling an armored superhero or droid, which transmits different emotions by changing the color of his eyes, the movement of his limbs and the tone of his voice
"The therapists say that engaging the child for a few seconds can be" Humans change their expressions in many ways, but the robots always do it in the same way, which is less frustrating, "said Rudovic
. for the child because the child learns in a very structured way how expressions will be shown. "
In addition to being personalized, the system created by these researchers involves a deep learning powered by AI, automatic speech recognition programs and objects, which makes it well suited to a problem such as understanding the multiple facets of the face, body and voice that help to understand a more abstract concept as the commitment of a child
Learning allows the robot to To directly extract the most important information from these data without humans having to manually shape them, "says Rudovic
For therapeutic robots, a personalized framework could be created from the data collected.
(This story was not edited by personal Devdiscourse and is generated automatically from a syndicated feed.)
Source link
