CONFERENCEThe aim of the module is to prepare high-level specialists in language technologies (LT), who are able to create solutions for LT problems and applications and are well prepared for the changing development of the industry, within the framework of bachelor's studies (BSP).
In order to achieve the goal, three main tasks are set, which students achieve by learning the courses included in the module:
1. To provide knowledge and understanding of natural language processing problems and their solution methods, so that students can independently use open source VT, conduct VT-related research, as well as develop new, practical VT solutions.
2. To provide knowledge of the Python programming language and develop modern programming skills, so that students can independently develop and integrate data processing systems, including use and integration of VT components.
3. To provide knowledge of the structure and use of deep machine learning solutions, develop practical skills in using deep machine learning frameworks, so that students can independently develop solutions for typical machine learning tasks - text and image classification, text and image processing.
Fundamentals of Language Technology (DatZB022)
Lecturers: professor, Dr.sc.comp. Inguna Skadiņa and professor, Dr.sc.comp. Normunds Grūzītis
Course language: Latvian and English
The aim of this course is to introduce students to the language technology and its use in practical applications and to provide knowledge about the fundamentals of natural language processing – the key issues and solutions.
The course covers the basic methods as well as the most important innovations and trends in the field of language technology. This includes language processing and modeling at different levels of text analysis by applying both knowledge-based and data-driven approaches.
The main focus is on data-driven methods and the language resources they require. We will primarily consider the aspects of processing the English and Latvian languages, using open-source libraries, language models and toolkits for practical solutions.
Course objectives:
- to introduce students to the main problems and fundamental methods of natural language processing;
- to widen students' knowledge and understanding about the importance and role of language technology in computer science and for society in general;
- to provide theoretical and practical knowledge for the use and integration of existing language technology solutions and for the development and evaluation of new language technology solutions.
Python programming language (DatZB084)
Lecturer: associate professor Dr.sc.comp. Uldis Bojārs
Course language: Latvian and English
The course aims to provide students with basic knowledge of the Python programming language.
Course objectives:
- to acquire basic knowledge and programming skills in the Python programming language;
- improve the programming skills of the participants of the course;
- become familiar with the available Python program packages and their use;
- acquire the main principles of problem solving using programming.
Fundamentals of Deep Machine Learning (DatZB056)
Lecturer: associate professor, Dr.sc.comp. Pēteris Paikens
Course language: Latvian
The aim of this course is to provide an overview of modern applications of machine learning and develop practical skills in using deep neural networks for common machine learning tasks - classification, text and image processing.
Course objective:
- to provide an introduction into artificial neural network based models;
- to provide an introduction to existing API frameworks for training such models.
Learning the course does not require prior knowledge of machine learning, but Python programming language skills are required, in which practical work will be carried out using the PyTorch framework.
Selected Topics in Deep Learning (DatZB130)
Lecturer: Dr. Maksims Ivanovs
Course language: Latvian
The aim of the course is to provide in-depth knowledge and practical skills in selected areas of deep learning, which are particularly relevant both in artificial intelligence research and in the development of solutions related to the use of artificial intelligence methods in the information technology sector:
- workflow for the development and use of deep neural network-based solutions;
- principles and best practices for adapting pre-trained deep neural network models;
- use of synthetic data for training machine learning models and evaluation of the quality of such data;
- explainable artificial intelligence methods.
Course objectives:
- to provide students with in-depth knowledge of modern deep learning methods and their applications in various data domains, building an understanding of typical workflow, related challenges, and their solutions;
- to develop students' ability to analyze and critically evaluate the selection of deep learning models, training and adaptation approaches, and their impact on model generalization ability and performance;
- to promote the ability to plan and implement reproducible experiments, taking into account data, computing resources, and time constraints as realistic system design factors;
- to strengthen skills of justifying and interpreting deep learning solutions, as well as communicating the achieved results to both professional audiences and the wider public.
To ensure that the module's study courses will achieve digital competence outcomes, the development process of study modules and courses was mapped and their compliance with the DigComp 2.2 framework was checked in several competence areas relevant to the Language Technology Initiative.