Aloha AI Network

How can we bring the future of AI to Hawaii’s students today? The Aloha AI Network and educational toolbox is our answer: a collaborative platform for students to develop AI and master Machine Learning

The Aloha AI Network is a collaborative infrastructure/service for students to use as they learn the fundamentals of AI machine learning, deep learning neural networks, and computer vision. Designed to be student-friendly, but industrial strength, the Aloha AI Network consists of three parts: Toolbox “edge” hardware sensors that are setup locally, in-depth training and eLearning tools from Oceanit experts, and an app utilizing cloud platform services to process captured data in a shareable, collaborative way. At the “edge” of Aloha AI, Oceanit provides AI-based IoT devices capable of simple image recognition computer vision functions; detecting & tracking objects such as people, dogs, cars, etc. Aloha AI’s edge system allows student users to learn about, develop, and build simple machine learning models for object detection.

In the cloud, the backend Aloha AI service collects data from the edge devices, consolidates it, and shares the data/insights among network subscribers such as different classes and student groups. These groups can then collaborate from across the state of Hawai’i on iterative models or shared projects using the AI toolbox. Student can build custom reports, create novel dashboards, build websites, and mobile applications using the toolbox.

Through a collaborative dashboard, students share their learnings and also build new applications for the deep learning models, together. Using image recognition, for example, students can train the Aloha AI to identify cats by analyzing example images that have been manually labelled as “cat” or “no cat”. Another example is using Aloha AI’s object detection to tally the numbers of vehicles at the school – ever wonder how many of your classmates drive trucks, SUVs , sedans, or mopeds? Aloha AI ingests thousands of examples from online sources and uses the student learning-model to optimize the detection of future cats.

With Aloha AI, students tap in to their creativity and design thinking processes to come up with ideas to use this AI data (e.g. keeping track of school library/cafeteria/parking usage, benchmark it against other schools, and so on). the objective of Aloha AI will be to establish a structure where students are collaborating to train their own models, develop novel uses, and build upon one another’s breakthroughs using the Aloha AI toolbox.

computer vision aloha ai oceanit 0