PB015 AI: the Practical Perspective

Fakulta informatiky
jaro 2026
Rozsah
2/0/0. 2 kr. (plus ukončení). Ukončení: zk.
Vyučováno kontaktně
Vyučující
doc. Mgr. Bc. Vít Nováček, PhD (přednášející)
Garance
doc. Mgr. Bc. Vít Nováček, PhD
Katedra strojového učení a zpracování dat – Fakulta informatiky
Dodavatelské pracoviště: Katedra strojového učení a zpracování dat – Fakulta informatiky
Předpoklady
IB111 (a soft prerequisite - a finished exam is not strictly required, but taking the course is strongly recommended for acquiring basic coding experience relevant to the course)
Omezení zápisu do předmětu
Předmět je nabízen i studentům mimo mateřské obory.
Mateřské obory/plány
Cíle předmětu
For decades, AI was mostly an academic sub-discipline of computer science with practical impact limited to narrow application scenarios the general public barely ever noticed. This has changed rather dramatically in recent years, though. Owing to the rapid advances in deep learning and large language model technologies, AI is now poised to disrupt the way we humans work, create, even live. Therefore, it is critical for professionals in virtually every field to understand AI at least at the intuitive level, even if they do not ever intend to actually write a single line of AI code.  This course aims at providing such an intuitive understanding. It will shed some light on what AI is (and what it isn’t), and provide broader historical and scientific context of the modern AI technologies. The course will also review a representative range of real world problems, showing where AI can help, and where it might best be avoided. Last but not least, we will discuss proper ways of talking about AI to overzealous managers or your neighbour panicking after hearing the latest evil superintelligence prophecy.
Výstupy z učení
- The graduates will gain a proper understanding of what AI is, what are its sub-disciplines and what has been their historical evolution up until the latest deep learning boom.
- The graduates will learn about the modern AI techniques and frameworks, and will be able to put them in a broader context of what has been motivating the AI developments over the years.
- The graduates will be able to select, apply and evaluate an appropriate AI approach to a variety of practical tasks across different data modalities.
- The graduates will be able to document, explain and justify their chosen approach to various stakeholders (product owners, sales or PR managers, collaborators, customers, neighbours, grandmothers, etc.).
Osnova
  • Motivations, historical overview (from Lovelace through Turing to Hinton et al.), classification of AI sub-disciplines
  • Search and optimisation 1/2 (state space search)
  • Search and optimisation 2/2 (local search)
  • Logic (basic formalisms such as propositional and predicate logic, AI-specific extensions such as situation calculus)
  • Knowledge representation (KR) and reasoning (semantic networks, ontologies, rules, knowledge graphs; automated reasoning engines); dealing with uncertainty (probabilistic and fuzzy extensions of logics, graphical models)
  • Machine learning (motivations, general overview, classification vs. regression, examples of supervised and unsupervised approaches)
  • Reinforcement learning (Q-learning, environments, policies, Markov decision processes, value function, direct policy search, Monte Carlo methods, gradient-based methods)
  • Artificial neural networks (motivations, general principles, examples of architectures)
  • Transformers and large language models (motivations, general principles, examples of architectures, prompt engineering)
  • Natural language processing (motivations, general overview, examples of specific use cases and techniques)
  • Image processing (motivations, general overview, examples of specific use cases and techniques)
  • Graph/networked data processing (motivations, general overview, examples of specific use cases and techniques)
  • Explainable AI, AI security, environmental, ethical and cultural concerns, future outlook of the field
Výukové metody
A weekly lecture.
Metody hodnocení
Final written exam as the baseline, compulsory evaluation (a mix of factual questions and essay-form assignments assessing the competence and creativity of students in using AI (or applicable alternatives) to solve samples of practical problems). Additional incentive points can be acquired for various tasks and activities during the semester (there is a minimum threshold of incentive points needed to pass the course). Examples of such tasks include successful completion of Python notebooks outlined in the lectures, elaboration of short essays on selected lecture topics or proactive participation in on-topic discussions during lectures.
Vyučovací jazyk
Angličtina
Další komentáře
Předmět je vyučován každoročně.
Výuka probíhá každý týden.
  • Statistika zápisu (nejnovější)
  • Permalink: https://is.muni.cz/predmet/fi/jaro2026/PB015