Course content

• What is machine learning?
  • How does machine learning differ from traditional programming? Correlation Vs Causation
  • Types of learning: Supervised, Unsupervised, Semi-Supervised and Reinforcement Learning
  • Difference between ML in research and in Production.
• Data Visualization using matplotlib and seaborn libraries
• Data preparation
  • How do I represent my data so that an algorithm can learn from it using the Pandas library?
  • Diversity and bias in data: How to identify that the prediction task is trained on representative data, Bias identification.
  • Feature engineering: Feature Selection and Feature Transformation.
• Machine learning Python libraries and practice
  • Machine learning libraries: Numpy, Pandas, Scikit-learn, Scipy, Tensorflow and Pytorch (for deep learning)
• Machine learning algorithms for Supervised and Unsupervised Learning
  • Linear and Logistic Regression, Decision trees, Support Vector Machines, K-means, K Nearest Neighbours, Dimensionality Reduction, Ensemble learning etc.
• Evaluation of results
  • Evaluation metrics. how to quantify "how bad" the prediction was? How do I increase the model's accuracy? Bias Variance tradeoff.
• AI ethics, responsibility, consequences
  • Case studies on model biases in the real world.
• Machine learning application practice in specialized domains, including:
  • Energy
  • Maritime
  • Medical

Learning outcome

Knowledge:

The candidate:

• Has a solid understanding of the fundamental concepts in machine learning, including supervised, unsupervised learning paradigms.
• Understands the mathematical foundations of key algorithms such as regression, classification, clustering, and neural networks.
• Has knowledge of model evaluation, bias-variance trade-off, overfitting/underfitting, and techniques for improving model generalization.
• Understands ethical and societal aspects of machine learning, including fairness, bias, and data privacy.

Skills

The candidate can:

• Pre-process and analyze real-world datasets using appropriate data transformation and feature-engineering techniques.
• Implement, train, and evaluate machine learning models using modern libraries and frameworks (e.g., scikit-learn, TensorFlow, PyTorch).
• Select and justify suitable learning algorithms for a given data problem based on data characteristics and project goals.
• Interpret model outputs and communicate findings clearly using visualization and performance metrics.
• Assess model robustness and limitations, and propose improvements or alternative methods.

General Competence

The candidate:

• Demonstrates an analytical and critical mindset toward data-driven decision-making and AI systems.
• Can reflect on the applicability, assumptions, and limitations of contemporary learning algorithms in different domains.
• Understands how to work collaboratively in interdisciplinary and project-based environments involving data, domains, and ethics.
• Recognizes the societal and ethical implications of AI applications and can discuss responsible use of machine learning technologies.
• Is prepared for lifelong learning in rapidly evolving Artificial Intelligence field and take advanced course on Deep Learning and Generative AI.

Learning methods and activities

Lectures for 4 hours along with exercises covering the entire course. Other supporting material (video, text, etc.) may be used. Mandatory assignments: 3 to be eligible for the oral exam.

Students can use AI based tools for correcting language of reports and taking help for better understanding but understanding is critical.

Compulsory assignments

• Assignments

Recommended previous knowledge

To understand the concepts presented and complete the exercise, we recommend that students meet the following prerequisites:

• Mastery of intro-level algebra: you should be comfortable with variables and coefficients, linear equations, graphs of functions. Being familiar with more advanced math concepts such derivatives is very helpful but not required.
• Proficiency in programming basics: it is highly recommended that you feel comfortable reading and writing Python code before starting this course. The programming exercise in the machine learning course uses Keras and pandas libraries.
• It is expected that Matlab programmers will find no problem to switch to Python.

Required previous knowledge

Recommended prior knowledge:

• Basic programming (Python).
• Statistics and probability.
• Linear algebra and calculus.
• Students lacking formal courses may be admitted if they can document equivalent skills.

Course materials

Reading list:

A. Geron. Hands-On Machine Learning with SciKit-Learn & TensorFlow: concepts, tools, & techniques to build intelligent systems. 2017, O'Reilly.

Materials, handouts, quizzes from various sources will be provided throughout the semester.

Credit reductions