Intelligent tutoring systems are shown to be effective in helping to show certain subjects, like algebra or grammar, but creating these computerized systems is difficult and laborious. Now, researchers at Carnegie Mellon University have shown they will rapidly build them by, in effect, teaching the pc to show.
Using a new method that employs AI, an educator can teach the pc by demonstrating several ways to unravel problems during a topic, like multicolumn addition, and correcting the pc if it responds incorrectly.
Notably, the pc system learns to not only solve the issues within the ways it had been taught, but also to generalize to unravel all other problems within the topic, and do so in ways in which might differ from those of the teacher, said Daniel Weitekamp III, a Ph.D. student in CMU’s Human-Computer Interaction Institute (HCII).
That challenge has been an unbroken problem for developers creating AI-based tutoring systems, said Ken Koedinger, professor of human-computer interaction and psychology. Intelligent tutoring systems are designed to continuously track student progress, provide next-step hints, and pick practice problems that help students learn new skills.
When Koedinger et al. began building the primary intelligent tutors, they programmed production rules by hand — a process, he said, that took about 200 hours of development for every hour of tutored instruction. Later, they might develop a shortcut, during which they might plan to demonstrate all possible ways of solving a drag. That cut development time to 40 or 50 hours, he noted, except for many topics, it’s practically impossible to demonstrate all possible solution paths for all possible problems, which reduces the shortcut’s applicability.
The new method may enable an educator to make a 30-minute lesson in about half-hour, which Koedinger termed “a grand vision” among developers of intelligent tutors.
A paper describing the tactic, authored by Weitekamp, Koedinger, and HCII System Scientist Erik Harpstead, was accepted by the Conference on Human Factors in Computing Systems (CHI 2020), which was scheduled for this month but canceled thanks to the COVID-19 pandemic. The paper has now been published within the conference proceedings within the Association for Computing Machinery’s Digital Library.
The new method makes use of a machine learning program that simulates how students learn. Weitekamp developed a teaching interface for this machine learning engine that’s user friendly and employs a “show-and-correct” process that’s much easier than programming.
For the CHI paper, the authors demonstrated their method on the subject of multicolumn addition, but the underlying machine learning engine has been shown to figure for a spread of subjects, including equation solving, fraction addition, chemistry, English grammar, and science experiment environments.
The method not only speeds the event of intelligent tutors but promises to form it possible for teachers, instead of AI programmers, to create their own computerized lessons. Some teachers, as an example, have their own preferences on how addition is taught, or which sort of notation to use in chemistry. The new interface could increase the adoption of intelligent tutors by enabling teachers to make the homework assignments they like for the AI tutor, Koedinger said.
Enabling teachers to create their own systems also could lead to deeper insights into learning, he added. The authoring process may help them recognize trouble spots for college kids that, as experts, they do not themselves encounter.