Course detail

Diploma Seminar (N-AIR)

FSI-VD8 Acad. year: 2027/2028 Summer semester

A master's thesis is no longer just proof that you can write a technical text. It is a small research project: you have to find where current solutions fall short, propose how to move them forward and convince the committee that your result holds up even under critical questioning.

The seminar builds on the bachelor's seminar and takes it a level higher. You will learn to map the literature systematically and identify the research gap, design the verification so that your result has something to be compared with, evaluate data with regard to uncertainty and write a discussion that neither overstates nor conceals. You will practise giving and receiving constructive feedback on a classmate's text, see how a master's thesis becomes a research paper, and find out how to use AI in research so that the results remain verifiable.

The semester culminates in a mock defence: a beta-presentation to which classmates respond with critical questions as a committee would. You will thus come to your defence with a thesis that has already withstood critical questioning once.

Language of instruction

Czech

Number of ECTS credits

4

Supervisor

Ing. Filip Kšica, Ph.D.

Department

Institute of Automation and Computer Science

Entry knowledge

The student has an assigned master's thesis topic and a basic overview of the field the thesis addresses. Experience with reading research papers in English and with writing a bachelor's thesis, i.e. working with sources, citing and structuring a text, is an advantage.

Rules for evaluation and completion of the course

The course is completed by course-unit credit.

Requirements for course-unit credit

  1. Active participation in seminars. Attendance is compulsory and recorded; two excused absences are tolerated, and any further absence must be compensated by completing the task of the missed seminar.
  2. Continuous assignments set in the e-learning system, for example peer feedback on a classmate's text and a response to it, or verifying the reproducibility of one's own result.
  3. Presentation of the research plan lasting 3 minutes: problem, gap, contribution and method of verification.
  4. Beta-presentation lasting 10–12 minutes followed by a discussion in which classmates ask questions as a committee would; it must include the research question, the method of verification and the results achieved so far.
  5. Poster summarising the master's thesis (problem, contribution, method of verification and main results), submitted to the e-learning system.

Presentations and assignments are submitted to the e-learning system. AI tools may be used in line with BUT principles; the student declares their use in the thesis and in submitted assignments.

Aims

A graduate of the course is able to:

  1. identify the research gap and the contribution of their master's thesis based on a systematic literature review,
  2. formulate a research question or hypothesis and design its verification, including comparison, metrics and success criteria,
  3. evaluate results with regard to uncertainty and repeatability and interpret them in relation to the literature without unfounded generalisations,
  4. critically assess a technical text against the assessment criteria for a master's thesis and give factual, constructive feedback,
  5. use AI tools in research transparently, verifiably and in line with ethical principles and BUT rules,
  6. defend the contribution of their thesis before the committee and respond factually to critical questions.

Study aids

The course e-learning provides seminar slides, assignments, assessment criteria for the master's thesis, a thesis template and links to BUT and FME directives.

The study programmes with the given course

Programme N-AIŘ-P: Applied Computer Science and Control, Master's, compulsory

Type of course unit

 

Exercise

26 hours, compulsory

Syllabus

1. The master's thesis as a research project: what changes compared with the bachelor's thesis, formalities and schedule, a work plan with milestones and risks; specifics of a thesis written in cooperation with a company.


2.–3. Finding the gap and defending the contribution: systematic literature review, mapping the literature and citation links, research question or hypothesis and success criteria.


4. Designing the verification: how to prove by experiment, simulation or design that the solution works; comparison, metrics, reproducibility and typical methodological errors such as data leakage or model overfitting.


5. Data that hold up: uncertainty, repeatability, statistical evaluation and graphs that actually say something.


6. Results and discussion: interpretation without overstatement, comparison with the literature and honestly acknowledged limitations.


7. Reading through the eyes of a reviewer: how a master's thesis is assessed, typical weaknesses and peer feedback on classmates' texts; how to respond to criticism factually.


8. AI in research: advanced tools for working with literature, data and code, verifiability of outputs, declaration of use and ethics of data and authorship; AI as a sparring partner when preparing for the committee's questions.


9. The research plan in three minutes, understandable even to a non-specialist: presentation of plans and review of peer feedback.


10. From thesis to paper and to defence: structure of a research paper, options for publishing results and the course of the defence.


11.–13. Mock defence: beta-presentations with discussion and committee-style questions; awarding of course-unit credits.