Artificial Intelligence and Machine Learning Syllabus

The following syllabus is representative. Universities may arrange subjects differently.

First-Year Subjects

Engineering Mathematics I: Calculus, matrices, vectors and differential equations.

Probability and Statistics: Probability rules, random variables, distributions and estimation.

Programming Fundamentals: Problem-solving, control structures, functions and debugging.

Python Programming: Data types, functions, classes, libraries and scientific computing.

C or C++ Programming: Memory, efficiency, structured programming and object-oriented concepts.

Engineering Physics: Scientific foundations relevant to computing and engineering.

Digital Logic: Boolean algebra, logic gates and digital circuits.

Engineering Design: Problem definition, prototyping and design communication.

Communication Skills: Technical writing, presentations and teamwork.

Second-Year Subjects

Data Structures: Arrays, linked lists, stacks, queues, trees, graphs and hashing.

Design and Analysis of Algorithms: Complexity, recursion, greedy methods, dynamic programming and graph algorithms.

Discrete Mathematics: Logic, sets, combinatorics, relations and graph theory.

Database Management Systems: SQL, relational design, transactions and database applications.

Operating Systems: Processes, memory, concurrency, storage and file systems.

Computer Networks: Network models, protocols, routing and internet applications.

Computer Organisation: Processor, memory, instructions and input-output systems.

Software Engineering: Requirements, architecture, testing, version control and maintenance.

Data Analysis and Visualisation: Data cleaning, exploration, charts and communication.

Third-Year Subjects

Artificial Intelligence Foundations: Intelligent agents, search, reasoning, planning and uncertainty.

Machine Learning: Regression, classification, clustering, model selection and evaluation.

Feature Engineering: Transformation, encoding, scaling, feature selection and dimensionality reduction.

Deep Learning: Neural networks, backpropagation, optimisation, regularisation and modern architectures.

Natural Language Processing: Text processing, embeddings, sequence models, transformers and language applications.

Computer Vision: Image processing, convolutional networks, detection and segmentation.

Big Data Systems: Distributed storage, parallel processing and large-scale analytics.

Cloud Computing: Virtualisation, scalable services and application deployment.

Data Mining: Pattern discovery, association, clustering and anomalies.

Final-Year Subjects

Reinforcement Learning: Agents, environments, rewards, policies and value functions.

Generative AI: Generative models, transformers, foundation models and multimodal applications.

MLOps: Model pipelines, deployment, versioning, monitoring and governance.

Explainable AI: Interpretation methods, feature attribution and communication.

Responsible AI: Fairness, accountability, transparency, privacy and safety.

AI Security: Data poisoning, adversarial inputs, model theft, prompt attacks and system protection.

Model Optimisation: Hyperparameter tuning, compression, quantisation and efficient inference.

Internship: Practical experience in software, data, AI or research.

Major Project: An end-to-end AI and ML system with evaluation and deployment.

Representative Semester-Wise AI and ML Syllabus

Core Machine Learning Subjects

Linear Regression: Predicting a continuous outcome through relationships among variables.

Logistic Regression: Estimating class probabilities for classification problems.

Decision Trees: Splitting data through interpretable decision rules.

Random Forests: Combining multiple trees to improve predictive stability.

Support Vector Machines: Constructing decision boundaries with maximum separation.

Clustering: Grouping similar observations without predefined labels.

Dimensionality Reduction: Representing high-dimensional data using fewer variables.

Neural Networks: Learning nonlinear relationships through connected computational units.

Ensemble Learning: Combining multiple models to improve performance.

Time-Series Learning: Modelling data ordered across time.

Model Evaluation

Students should understand:

• training, validation and test sets;
• cross-validation;
• accuracy;
• precision;
• recall;
• F1 score;
• ROC-AUC;
• mean absolute error;
• root mean square error;
• confusion matrices;
• calibration;
• class imbalance;
• overfitting;
• underfitting;
• data leakage;
• baseline models; and
• statistical uncertainty.

Selecting the wrong metric can produce a model that looks successful but fails the actual objective.

Data Preparation and Feature Engineering

Data preparation may include:

• missing-value handling;
• duplicate removal;
• outlier assessment;
• categorical encoding;
• numerical scaling;
• text tokenisation;
• image augmentation;
• feature selection;
• dimensionality reduction;
• label quality checks; and
• train-test separation.

Data preparation often requires more time than model training.

Deep Learning Subjects

Deep-learning study may cover:

• perceptrons;
• feedforward networks;
• activation functions;
• loss functions;
• backpropagation;
• gradient descent;
• convolutional networks;
• recurrent networks;
• attention;
• transformers;
• autoencoders;
• generative adversarial networks;
• regularisation; and
• transfer learning.

Natural Language Processing Subjects

Students may learn:

• text preprocessing;
• language representation;
• sentiment analysis;
• information extraction;
• named-entity recognition;
• text classification;
• question answering;
• translation;
• summarisation;
• language models;
• retrieval-augmented generation; and
• NLP evaluation.

Computer Vision Subjects

Students may study:

• image representation;
• filtering;
• edges;
• features;
• image classification;
• object detection;
• segmentation;
• pose estimation;
• video understanding;
• transfer learning; and
• vision-model evaluation.

AI and ML Electives

Possible electives include:

• probabilistic graphical models;
• advanced reinforcement learning;
• graph neural networks;
• speech technology;
• recommender systems;
• autonomous systems;
• robotics;
• healthcare AI;
• financial machine learning;
• cyber-security analytics;
• causal inference;
• federated learning;
• privacy-preserving ML;
• multimodal AI;
• foundation models;
• edge AI;
• AI product management;
• human-centred AI;
• AI policy and governance; and
• high-performance computing.

AI and ML Laboratories

Python Laboratory: Programming and scientific-computing exercises.

Data Structures Laboratory: Implementation and analysis of core structures.

Data Analytics Laboratory: Cleaning, visualisation and statistical analysis.

Machine Learning Laboratory: Training and comparison of ML algorithms.

Deep Learning Laboratory: Neural-network implementation and experimentation.

NLP Laboratory: Text classification, language models and information extraction.

Computer Vision Laboratory: Image classification, detection and segmentation.

MLOps Laboratory: Packaging, deployment, versioning, monitoring and automation.

Responsible AI Laboratory: Bias, explainability, privacy and robustness assessment.

AI and ML Project Ideas

Students may build:

• customer-churn prediction;
• credit-risk support model;
• crop-yield forecasting;
• plant-disease recognition;
• manufacturing-defect detection;
• predictive-maintenance system;
• personalised recommendation engine;
• multilingual document classifier;
• speech-command recognition;
• traffic-congestion forecasting;
• energy-demand prediction;
• fake-review analysis;
• medical-image triage support;
• student-learning analytics;
• inventory forecasting;
• phishing detection;
• network-anomaly detection;
• explainable fraud detection;
• automated document search;
• accessible image-description tool;
• content-moderation support system;
• time-series forecasting platform;
• small retrieval-augmented assistant;
• model-drift monitoring system; or
• privacy-preserving learning experiment.

Projects should use data legally and explain limitations clearly.

Capstone Project Requirements

A strong capstone should include:

• problem statement;
• target users;
• data source;
• legal and ethical assessment;
• exploratory analysis;
• baseline;
• model comparison;
• evaluation metric;
• error analysis;
• fairness assessment;
• deployment;
• monitoring plan;
• documentation; and
• demonstration.