Syllabus

Introduction to ML: AI vs ML vs DL, ML pipeline, datasets, bias-variance, Overfitting and Underfitting,

Data Cleaning, Missing Value Treatment, Outlier Detection, Data Transformation, Feature Scaling, Normalization, Standardization, Label Encoding, One-Hot Encoding, Feature Selection, Feature Extraction, Principal Component Analysis (PCA), Data Visualization, Exploratory Data Analysis (EDA)

Supervised Learning: Regression, Linear Regression Multiple Linear Regression Polynomial Regression Ridge Regression Lasso Regression Elastic Net
Classification: Logistic Regression K-Nearest Neighbors (KNN) Naïve Bayes Decision Trees Random Forest Support Vector Machine (SVM) Ensemble Learning Gradient Boosting XGBoost (Introduction)

Unsupervised Learning: Clustering K-Means Hierarchical Clustering DBSCAN Gaussian Mixture Models Dimensionality Reduction PCA t-SNE (Introduction) Association Rule Mining Apriori Algorithm Market Basket Analysis Anomaly Detection

Advanced Machine Learning Practices: Explainable AI (XAI) SHAP LIME Responsible AI Fairness and Bias AutoML Model Deployment REST APIs Model Serialization (Pickle, Joblib) MLOps Overview ML Lifecycle Case Studies in: Healthcare Finance Smart Agriculture Computer Vision etc.

Understand the mathematical foundations, concepts, and paradigms of machine learning.

Develop, train, evaluate, and optimize supervised and unsupervised machine learning models.

Apply machine learning techniques to solve real-world problems using Python and industry-standard libraries.

Analyze model performance using appropriate evaluation metrics and validation strategies.

Design end-to-end machine learning solutions incorporating feature engineering, model deployment, and ethical AI considerations.

Explain the principles, mathematical foundations, and workflow of machine learning algorithms.

Apply supervised learning algorithms for classification and regression problems using appropriate preprocessing techniques.

Implement unsupervised learning algorithms for clustering, dimensionality reduction, and anomaly detection.

Evaluate, compare, and optimize machine learning models using performance metrics, cross-validation, and hyperparameter tuning.

Design complete machine learning solutions for real-world applications considering deployment, scalability, explainability, fairness, and ethics.

Tom Mitchell, Machine Learning, TMH

M. Bishop, Pattern Recognition and Machine Learning, Springer