Adaptive Learning

Since of the emergence of artificial intelligence (AI), adaptive learn, a path that is customarily engineered to each student, is no long a dream. A edTech start up called Knewton (https://www.knewton.com/), Founded in 2008 at New York City, had pioneered to help students, served over 13 million students from all over the world, through their unique learning history to improve their future learning. That is huge. Think of our traditional boring class, in which students are yawning and sleeping while the annoying teacher still stand murmuring in the platform, and the scenes are everywhere. Our education system was designed for the perspective of producers, i.e. teachers, rather than customers, i.e. students. Therefore, most of the time that spent in the class is mostly inefficient and useless. Students are blames to be not intelligent enough if they are not graded A under such system. Ain’t we supposed to blame the educators’ teaching method or the mistaken educational system?

I really appreciate the assumption from Knewton that “no two student come from the same background or learn the same way.” Therefore, our education system should offer adaptive courseware for every student based on evolution of each learning history. For example, they build the software that is used to track students’ progress through online education system. Student can learn the their course work step by step, just like playing video games. As a professor or teacher, he/she could understand the students’ acceptance to each chapter and constitute more reasonable and responsive teaching materials. As a student, he/she can also utilize this software to help to relate his/her past experiences and new materials.

The emergence of deep learning also expands the utilization on adaptive learning. We can regard each node/kernel in deep learning as a state, which is mapped to a state for specific learner. For example, a kernel can represent the responsive of the score as one student study Calculus within his/her class. It also can used to represent someone’s sensitivity to specific chapter for specific subject at the specific age. The mapping function is very complicated that no standard model can be built since it also needs to consider not only physical but psychological sense. In addition, data analytics can also play a big role for analyzing the validity of the measure model and future trends as a teaching recommendation system for educators.

The AI for our education is huge. The edTech can revolutionize our whole education system to be more customized and efficient through adaptive learning. Also, since the popularity of online courses, teachers do not need to be familiarity with all the course content. On the contrary, he/she is expected to offer what they are really good at. Students can also beneficial though this online course-ware though dedicated grading system, adjustable content provider, and flexible to access the learning system anywhere any time. It is because everyone is born to be uniqueness and the AI makes our system more humanity, especially for edTech. And the expectation is that we were born to learn and the revolutionized could help us to learn by nature.

Like what you read? Give Ming-Chia Tsai a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.