This chapter first introduces the foundational philosophy and structure of machine learning Learningmachine learning through the lens of scientific modeling. It emphasizes that, like scientific theories, machine learningLearningmachine learning models are surrogates of reality, designed to both explain existing data and predict future outcomes. The machine learningLearningmachine learning process typically involves three core components: defining a hypothesisHypothesis classClass (modeling assumption), choosing a scoring mechanism (loss function)Function, and selecting a searchSearch strategy (optimization). These elements collectively form a general workflow for learning from data, reflecting the essential trade-off between model fit and generalizationGeneralization. The chapter continues by defining machine learningLearningmachine learning and its key fundamental tasks—classificationClassification, regressionRegression, dimensionality reductionDimensiondimensionality reduction, and reinforcement learningLearningreinforcement learning. It then surveys essential methods in statisticalStatistical learning and concludes with a historical overview of neural networksNeural network and deepLearningdeep learning learningDeepdeep learning, situating them within the broader context of machine learningLearningmachine learning.