Non-convex Optimization for Machine Learning explores how to solve complex problems in machine learning using non-convex optimization techniques. The focus is on providing practical tools and methods for tackling these challenges. The research covers basic concepts in convex analysis and optimization, then delves into their non-convex counterparts. The study also examines real-world applications in machine learning and signal processing, showing how these techniques can be applied effectively. The monograph includes exercises and references for further reading, making it a valuable resource for courses in optimization, machine learning, and signal processing.