Approachable Machine Learning Education: A Spiral Pedagogy Approach with Experiential Learning

Machine learning (ML) is an important subject for computer science students to learn due to its broad applications. Introductory courses often present techniques in a linear sequence, resulting in a steep learning curve that can overwhelm students and limit the time for experiential learning through course projects. To address this, I restructured the course using a spiral approach, presenting concepts in three iterations. Each iteration delves deeper into the material and introduces complex computational topics progressively. This method includes a built-in repetition mechanism that reinforces learning and enhances understanding. Moreover, this approach allows time for hands-on projects that apply theory to real-world scenarios, helping students better understand the course materials. The spiral approach was implemented in an ML course at a local university, resulting in positive student feedback and improved course retention rates.