AI in Reopening Schools
Published by Sid Chadha on August 7th, 2020
In today's world, it is vital to provide education to every student. Though during this global pandemic, the question of how to do this is a popular debate. One of the most essential tools that we are seeing being implemented in school re-openings all over the world is artificial intelligence, which can be utilized in both in-person and distance learning.
In the Case of In-Person Learning
This is perhaps one of the most advanced components of Artificial intelligence being used in re-opening schools. This temperature scanner can read the temperature of multiple parts of a human body. Numerous primary schools in China are using this tool to combat the lack of detection within schools. These AI thermal scanners can passively measure up to 3000's students body's temperature. This unprecedented AI scanner is supported by a NPU (natural processing unit - a neural network that functions similar to the brain) that can calculate at unprecedented rates (3 trillion times per second) with its algorithm. This works through using an infrared distance compression algorithm (compressing data to store while scanning someones temperature through an infrared sensor) and body calibration software to maintain almost a 0.3 degree accuracy. This scanner is now not only being used in schools all over China, but also train stations, airports, and hospitals in China, with inquiries for these sensors from countries all over the world. The algorithm behind the infrared sensor could work through using a simple convolutional neural network being powered by a powerful compression algorithm, being able to compress data easily, so that this algorithm works at maximum efficiency
Illustration of Firetas's scanners; buisnesswire
But it's not only thermal detection that these sensors could come in use for. They could also detect if masks are being worn. Through using another convolutional neural network, it would be easy to detect if masks are being worn over someones mouth and nose. Through implementing AI in a mask sensor, it would be easy to mitigate the spread of the virus within schools
Mask Detection software; Scylla AI
A sensor could additionally be used to recognize if social distancing is taking place, additionally through a convolutional neural network. The company that has been investigating this technology uses artificial intelligence to help businesses, schools, housing complexes and other facilities combat the coronavirus by tapping into surveillance cameras and identifying when and where people are getting too close together and for how long.
Social Distancing AI System; Scylla AI
Automatic disinfection spray
By using convolutional neural networks to recognize where to disinfect an area, this robot can automatically disinfect any coronavirus particles. This robot would be extremely useful by cleaning the air and surfaces.
Picture of a AI enabled robot able to clean surfaces automatically; YouTube
In the Case of Online School
If most schools go online, Artificial Intelligence can also be implemented here too. One popular application is using a deep feed forward neural network to scan for inappropriate language. Even though this technology is used in China to oppress certain sets of language, it can work as extremely effective in chats for online school currently taking place there.
Companies such as Microsoft have also worked on something similar but in actual voice chats. By using a convolutional neural network with multiple layers geared towards audio, they can censor what people are actually saying so that no one can use profane language. Microsoft put out a statement that said: "Ultimately our vision is to supplement our existing efforts and leverage our company efforts in AI and machine learning technology to provide filtration across all types of content on Xbox Live, delivering control to each and every individual player. Your feedback is more important than ever as we continue to evolve this experience and make Xbox a safe, welcome and inclusive place to game", displaying their ambition to implement AI censoring. This could also be implemented in school settings in virtual synchronous learning.
In terms of online education, Squirrel AI is leading the charge in using online AI in virtual education platforms. Founded in 2014, Squirrel is “the first pure-play AI-powered adaptive education provider in China”.
Squirrel works through partnering with highly-skilled teachers and divides given classes into the extremely small conceptual blocks of material. As an example, Squirrel can break down middle school-level math into above 10,000 knowledge points . This means that Squirrel can help read how skilled a student is at each subtopic at an extremely high accuracy.
Squirrel AI – image by Noah Sheldon for MIT Technology Review
These types of algorithms have already began to be implemented in the United States. Karen Hao in her MIT Technology Review article furthers “By comparison, a textbook might divide the same subject into 3,000 points; ALEKS, an adaptive learning platform developed by US-based McGraw-Hill, which inspired Squirrel’s, divides it into roughly 1,000.”
In Sweden, Sana Labs takes a different approach in AI for education through helping other companies personalize education by taking care of the AI and machine learning side of things.
Sana covers several fields: mathematics, language learning, and professional education. For a Southeast Asian education company, Sana was able to build a personalized homework system. For another customer in finance exam preparation, Sana provided personalized review sessions that was able to suggest optimized next steps for each individual student. These customer case studies from Sana Labs demonstrate an ALO7-like approach: using AI to free up teachers’ time and personalize education without removing the human element.
Overall, if schools open up in person or online, AI will be readily available to ease up the job of administrators, teachers, and cleaning staff.