Veidlapa Nr. M-3 (8)

Study Course Description

Advanced Machine Learning

Main Study Course Information

Course Code

SZF_168

Branch of Science

Electrical engineering, Electronic engineering, Information engineering; Other Sub-Branches of Electrical Engineering, Electronics, Information and Communication Technology

ECTS

5.00

Target Audience

Business Management; Management Science

LQF

Level 7

Study Type And Form

Full-Time; Part-Time

Study Course Implementer

Course Supervisor

Nataliia Kinash

Structure Unit Manager

Ieva Puzo

Structural Unit

Faculty of Social Sciences

Contacts

Kuldīgas 9c, Rīga

About Study Course

Objective

By the end of this course, students will learn to apply advanced machine learning techniques, including neural networks, LLMs, reinforcement learning, GANs, and autoencoders, to solve business problems. They will learn to build, deploy, monitor, and explain models, ensuring ethical use and strategic business impact.

Preliminary Knowledge

"Fundamentals of artificial intelligence and machine learning" course or equivalent.

Learning Outcomes

Knowledge

1.Explain advanced machine learning concepts, models, and algorithms, including their theoretical foundations and practical implications.

Individual work and tests

Individual work • Literature studies

2.Describe suitable use-cases for advanced neural network architectures such as CNNs, RNNs, GANs, and LLMs, including their advantages, limitations, and suitable applications.

Individual work and tests

Literature studies • Individual work

3.Describe the fundamental principles of reinforcement learning and differentiate it from supervised and unsupervised learning approaches.

Individual work and tests

Individual work • Literature studies

4.Discuss ethical considerations in AI, including algorithmic bias, fairness, transparency, and accountability.

Individual work and tests

Literature studies • Individual work

5.Outline the key steps in model deployment and explain best practices for monitoring and maintaining model performance in production environments.

Individual work and tests

Individual work • Literature studies

Skills

1.Build and fine-tune advanced machine learning models

Individual work and tests

Independent assignments

2.Use Natural Language Processing Techniques including prompt engineering.

Individual work and tests

Independent assignments

3.Deploy machine learning models.

Individual work and tests

Independent assignments

4.Use explainable AI techniques like SHAP and LIME for model interpretability.

Individual work and tests

Independent assignments

5.Choose the appropriate advanced neural network architecture such as CNNs, RNNs, GANs, and LLMs for the business-case.

Individual work and tests

Independent assignments

Competences

1.Choose and apply machine learning solutions to improve business decision-making and efficiency.

Individual work and tests

Presentation of a deep learning solution

2.Ensure ethical AI practices are integrated.

Individual work and tests

Presentation of a deep learning solution

3.Adapt machine learning models in dynamic environments.

Individual work and tests

Presentation of a deep learning solution

4.Evaluate performance of deep neural networks with appropriate metrics and interpret the results.

Individual work and tests

Presentation of a deep learning solution

5.Communicate model results and insights effectively to non-technical stakeholders.

Individual work and tests

Presentation of a deep learning solution

Assessment

Individual work

Title	% from total grade	Grade
1. Individual work	15.00% from total grade	10 points
After each lecture studens complete and online quiz assessing the understanding of theoretical material covered in the lecture.
2. Literature studies	-	-
To prepare for lectures and practical classes students should read the assigned literature and watch the video materials available on the Moodle learning platform.

Examination

Title	% from total grade	Grade
1. Independent assignments	35.00% from total grade	10 points
Short independent assignments completed during the course will be evaluated. The assignments consist of practical in-class exercises. Assignments focus on applying advanced ML techniques in business scenarios.
2. Presentation of a deep learning solution	50.00% from total grade	10 points
In the final class, each student will present their project, demonstrating their understanding and practical application of key concepts. The final project integrates multiple concepts, requiring students to solve a real-world business problem and present results. The project should be submitted in advance according to the schedule published on the e-larning platform (Moodle). The presentation will consist of 10 minute demonstration that can includes slides (if necessary), a working demo, and code. Students will demonstrate their understanding of the solution by answering questions from the lecturer and peers.

Study Course Theme Plan

FULL-TIME

Part 1

Lecture

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Advanced Machine Learning

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Advanced Feature Engineering and Data Transformation. Model Evaluation and Selection.

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Neural Networks

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Regularization and Optimization in Neural Networks

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Convolutional Neural Networks (CNNs) for Image Analysis

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Advanced CNN Architectures and Transfer Learning

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Time Series Analysis with Recurrenct Neural Networks (RNNs and LSTMs)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Natural Language Processing with Transformers

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Large Language Models (LLMs)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Prompt Engineering for LLMs

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Generative Adversarial Networks (GANs) and Synthetic Data

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Autoencoders and Anomaly Detection

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Reinforcement Learning (RL)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Advanced Reinforcement Learning (Policy-Based Methods)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Meta-Learning and Adaptive Models

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Explainable AI and Interpretability

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Ethics and Responsible AI for Customer-Facing Algorithms

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Model Deployment Strategies and Business Integrations

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Model Monitoring and Maintenance

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Project Presentation and Wrap-up

Total ECTS (Creditpoints):

5.00

Contact hours:

40 Academic Hours

Final Examination:

Exam (Written)

PART-TIME

Part 1

Lecture

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Advanced Machine Learning

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Advanced Feature Engineering and Data Transformation. Model Evaluation and Selection.

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Neural Networks

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Regularization and Optimization in Neural Networks

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Convolutional Neural Networks (CNNs) for Image Analysis

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Advanced CNN Architectures and Transfer Learning

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Time Series Analysis with Recurrenct Neural Networks (RNNs and LSTMs)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Natural Language Processing with Transformers

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Large Language Models (LLMs)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Prompt Engineering for LLMs

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Generative Adversarial Networks (GANs) and Synthetic Data

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Autoencoders and Anomaly Detection

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Introduction to Reinforcement Learning (RL)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Advanced Reinforcement Learning (Policy-Based Methods)

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Meta-Learning and Adaptive Models

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Explainable AI and Interpretability

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Ethics and Responsible AI for Customer-Facing Algorithms

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Model Deployment Strategies and Business Integrations

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Model Monitoring and Maintenance

Class/Seminar

Modality	Location	Contact hours
On site	Auditorium	2

Topics

Project Presentation and Wrap-up

Total ECTS (Creditpoints):

5.00

Contact hours:

40 Academic Hours

Final Examination:

Exam (Written)

Bibliography

Required Reading

1.

Raschka, S., Liu, Y., Mirjalili, V. "Machine Learning with PyTorch and Scikit-Learn"Suitable for English stream

2.

Foster, D., "Generative Deep Learning"

Additional Reading

1.

Goodfellow, I., Bengio, Y., & Courville, A. "Deep Learning"Suitable for English stream

2.

Barocas, S., Hardt, M., & Narayanan, A. "Fairness and Machine Learning: Limitations and Opportunities"Suitable for English stream

3.

Molnar, C. "Interpretable Machine Learning"Suitable for English stream

4.

Géron, A. "Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow", 2nd EditionSuitable for English stream

5.

Sutton, R. S., & Barto, A. G. "Reinforcement Learning: An Introduction"Suitable for English stream

6.

Bishop, C. "Pattern Recognition and Machine Learning"Suitable for English stream

7.

Rudin, C., Chen, C., Chen, Z., Huang, H., Semenova, L., Zhong, C. “Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges” (Statistics Surveys, 2021).

8.

Goodfellow, I. J., Pouget‑Abadie, J., Mirza, M., Xu, B., Warde‑Farley, D., Ozair, S., Courville, A., Bengio, Y. “Generative Adversarial Networks” (arXiv preprint, 2014).

9.

Iusztin, P., Labonne, M. "LLM Engineer's Handbook: Master the art of engineering large language models from concept to production"

10.

Noyan, M., Marafioti, A., Farré, M., Zohar, O. "Vision Language Models"

Other Information Sources

1.

The Transformers by Hugging Face (open-source, website and documentation)Suitable for English stream