Author Description

Machine learning, a subset of artificial intelligence, is one of the most transformative technologies of our time. And yet, machine learning and data exploration are minimally covered in the standard school curriculum. Machine Learning With School Level Math brings you a compelling program with a brand-new approach. Our stance is that many elements of machine learning are completely accessible without the sophistication of advanced math and within the reach of middle to high school students. All they need is the math they are already learning in school. Developed collaboratively by a physicist with significant teaching experience including K-12 teaching experience and an algorithm development industry expert, our program is a first-of-its-kind effort to demystify the inner workings of machine learning at the school level. Quick highlights: Prerequisites: No advanced math or programming background is assumed. 8 chapters, 30 worksheets, and more than 100 fully worked-out pen-and-paper and computer problems. Learn Python programming with unique and innovative CodeTrailMaps From concepts of rate and slope, the gradient descent algorithm (GDA) is developed, and then the GDA is used to solve linear regression, logistic regression, and neural nets. Composite functions, part of current school course structures, are used to introduce neural nets. Through these models, students learn a complete machine learning workflow, including data normalization, model selection, finding model parameters using a pertinent cost function, train/test cycles, and the construction of confusion matrices to evaluate model performance. Ideas of dimensionality reduction with principal component analysis are also covered as one more application of GDA. Basic Python programming is introduced seamlessly alongside the theory to deliver a thorough hands-on experience. A general audience, curious about the fundamentals of machine learning will also find value in this book.