Deep Learning Manual - Paul Stephen Prueitt - Bog

This latest volume in the Deep Learning Manual series presents an architecture and roadmap for implementing, on a nationwide basis, a deep learning-based education platform. The proposed platform combines adaptive learning software and behavioral neuroscience in a unique and powerful way. The author's motivation for this approach comes from decades of observing classroom behaviors that limit rather than advance our individual and collective capacity for self-discovery. Many of these behaviors reinforce longstanding inequalities. Education need not be difficult in our current age, regardless of a person's socio-economic background. Much of the basis for students' struggles, especially in college-level mathematics, is the over-emphasis on rote memorization at the expense of understanding the topics in ways that make sense to the individual student. For this reason, sense-making tools and practices form a key aspect of the proposed platform. Prueitt's approach is based on the premise of having a "sense of location" as a student navigates a "knowledge space" of topics, moving from those that are "comfortable" to those that are "uncomfortable." This book paints both a vision and a roadmap to the future, where first principles of science and mathematics are communicated through a "learning medium" that builds upon traditional books and research publications. The goal is to create a national and even global network of knowledge media that improves an average citizen's capacity for lifelong learning. Design Principles is richly illustrated with anonymized images of real handwritten narratives from freshman college mathematics students who have gone from struggling with and disdain toward the subject to feeling knowledgeable and empowered to apply the learned principles in future classes, work, and life.