Language technologies module MSP

The aim of the Language Technology (LT) module is to prepare high-level specialists in the field of LT (including deep machine learning) within the framework of the Master's study programme (MSP), who are able to integrate existing and create new solutions for diverse, practical LT tasks and applications and are well prepared for the dynamic development of the sector. 

Credit points: 9 ECTS

Study courses included in the study module:

1. DatZM037 “Applications of Language Technology” 3 ECTS
2. DatZD025 “Dziļā mašīnmācīšanās” 6 ECTS

In order to achieve the goal, two main tasks are set, which students achieve by learning the courses included in the module:

1. To provide theoretical knowledge and practical skills for working with modern multilingual and multimodal technologies for language, including multimodal content, understanding and language generation, speech recognition and speech synthesis, so that students can independently use and integrate language technologies in ICT solutions and process digitalization, as well as create new and innovative solutions (DatZM037).

2. To provide in-depth knowledge of the concepts and methods of supervised and unsupervised machine learning: from text meaning coding (MLM – Masked Language Models) to large, generative language and multimodal models (LLM – Large Language Models), so that students can understand and independently solve the most complex tasks using these modern artificial intelligence methods (DatZD025).

Applications of Language Technology (DatZM037)

Lecturer: Dr.sc.comp., assoc. prof. Normunds Grūzītis

Credit points: 3 ECTS

Time of lectures: Thursdays 16:30 until 18:00.

The aim of the course is to learn the latest language technologies for processing text and speech, including multimodal data, and to be able to use these technologies in the development of practical applications. The course will cover various types of transformer models for solving various types of tasks.
The main focus will be on multilingual models, their use, evaluation and adaptation capabilities. Open source language models and programming frameworks will be used to acquire practical skills. The main methods and solutions for the combined use of generative language models and knowledge bases/graphs will also be covered.

Deep Machine Learning (DatZD025)

Lecturer: Dr.sc.comp., prof. Guntis Bārzdiņš

Credit points: 6 ECTS

The course aims to examine deep neural machine learning methods, which allow predicting future results based on available past training data. Machine learning is a cornerstone in Data Science, Big Data Analytics, Robotics, Natural Language Processing and other areas of artificial intelligence. As computers and GPUs become increasingly powerful, deep neural networks have gradually replaced the simplest Machine Learning methods. The course aims to introduce students to the concepts of supervised and unsupervised machine learning from primitive classification to large language models and Generative Artificial Intelligence. No prior knowledge of machine learning is required to complete the course, but general programming skills are required for homework.