This course provides a comprehensive foundation in Artificial Intelligence and Machine Learning by integrating theoretical principles, computational methods, and practical applications. Learners are introduced to the evolution of AI concepts, data-driven problem solving, machine learning paradigms, and modern deep learning architectures. Through progressive exposure to Python-based AI tools, statistical learning techniques, linear and nonlinear algorithms, ensemble methods, and neural network models, the course develops both conceptual understanding and technical competence. Emphasis is placed on ethical considerations, model evaluation, generalization, and real-world implementation in domains such as computer vision, natural language processing, and intelligent systems development.

CLO 1: Explain core principles, paradigms, and ethical considerations of Artificial Intelligence and Machine Learning.

CLO 2: Apply Python-based AI and machine learning libraries to preprocess data, build models, and visualize analytical results.

CLO 3: Evaluate and compare linear, nonlinear, ensemble, and deep learning algorithms for different problem domains.

CLO 4: Design and implement intelligent systems using modern machine learning and deep learning frameworks for practical applications.

• Review of Artificial Intelligence Concepts — Revisits foundational AI principles: intelligent agents, problem-space search, heuristic methods, knowledge representation, and symbolic reasoning. Introduces ethical and societal perspectives.
  
  • Establishes conceptual grounding and continuity from undergraduate AI, setting the stage for advanced study.
• Python Foundations for Artificial Intelligence — Covers Python programming environments for AI including Numpy, Pandas, Scikit-learn, Matplotlib, and TensorFlow setup. Includes code structure, data handling, and visualization.
  
  • Introduced early so learners can apply computational skills throughout subsequent modules.
• Data in Machine Learning — Explores statistical and computational perspectives of data: types, preprocessing, normalization, feature extraction, dimensionality reduction, and handling bias.
  
  • Ensures data literacy before learners engage with algorithmic models.
• Learning and Generalization in Artificial Intelligence — Introduces mapping from input to output, learning functions, bias-variance trade-off, underfitting/overfitting, model generalization, and regularization.
  
  • Provides the conceptual bridge between data and algorithms while establishing theoretical foundations of model behavior.
• Learning Paradigms: Supervised, Unsupervised, and Semi-Supervised Learning — Compares the major machine learning paradigms, task categories, training objectives, and representative algorithms.
  
  • Provides an essential taxonomy of learning approaches before studying specific algorithm families.
• Linear Algorithms for Learning — Examines Linear Regression, Logistic Regression, Linear Discriminant Analysis (LDA), and Gradient Descent optimization techniques. Covers evaluation metrics and model interpretability.
  
  • Builds from theoretical understanding toward mathematically simple but effective learning models.
• Nonlinear Algorithms for Learning — Covers Decision Trees, Naïve Bayes, k-Nearest Neighbour (kNN), Support Vector Machines (SVM), and Learning Vector Quantization (LVQ). Discusses kernel tricks and nonlinear separability.
  
  • Extends learning from linear models to complex decision boundaries and richer algorithmic approaches.
• Ensemble and Hybrid Learning Algorithms — Focuses on Bagging, Random Forests, Boosting (AdaBoost, Gradient Boosting, XGBoost), and Stacking. Includes bias–variance reduction and model fusion concepts.
  
  • Acts as a transition between conventional machine learning and deep learning through meta-learning strategies.
• Deep Learning I: Foundations and Architectures — Introduces perceptrons, activation functions, loss functions, backpropagation, gradient-based optimization, and multilayer perceptrons (MLPs).
  
  • Marks the transition to representation learning and establishes the mathematical basis of deep learning.
• Deep Learning II: Advanced Models and Applications — Covers CNNs for image recognition, RNNs/LSTMs for sequential data, transfer learning, and frameworks such as Keras and TensorFlow. Includes case studies in computer vision, NLP, and education.
  
  • Integrates prior knowledge into practical, high-impact artificial intelligence applications.