Constraint-Based Solution Methods for Automated Production Scheduling

Modern-day factories often employ complex and highly automated processes to manufacture large quantities of products. Consequently, devising an efficient production schedule presents a challenge, as numerous complex constraints typically impose restrictions on feasible schedules. To capture the requirements of production scheduling problems across various sectors, high-level constraint modelling languages – which can be used in conjunction with automated constrained optimisation solving technology – have been proposed in the past. However, to handle large-scale scheduling problems efficiently, novel modelling techniques and intelligent algorithms are often required to tackle the complex scheduling constraints found in real-world environments.

This talk will provide an overview of our work on constraint-based solution approaches for several real-world scheduling problems from industry. We will discuss advanced constraint modelling strategies and state-of-the-art solution algorithms.

Understanding Large Language Models and the Path to AGI

Neural networks are often perceived by the general public as a form of magic, but at their core, they are essentially a structured sequence of mathematical transformations mapping an input tensor space to an output tensor space. Large Language Models (LLMs), such as ChatGPT, operate through a series of tensor algebra operations, utilising vast amounts of data and computational power. The true ‘magic’ arises not from individual calculations but from scaling—as models grow larger, they exhibit emergent properties that were not explicitly programmed.

This talk explores the implications of scale in AI, drawing lessons from nature. Evolution did not endow humans with 86 billion neurons and 100 trillion synaptic connections by chance; nature is economical, and the complexity of human intelligence is deeply linked to its capacity. The human brain’s encephalisation quotient—the ratio of brain mass to body size—exceeds that of any other primate, highlighting the importance of scale in biological intelligence.

A central question arises: Are human intelligence and consciousness Turing-computable? If intelligence is simply the product of sufficient capacity and complexity, then in principle, AI models, when scaled up, should be able to achieve human-level Artificial General Intelligence (AGI). But does the nature of intelligence go beyond computation? What is the Kolmogorov complexity of human intelligence? The Chinese Room argument, proposed by the philosopher John Searle, challenges the idea that syntactic manipulation alone is sufficient for true understanding. Meanwhile, philosopher and cognitive scientist Daniel Dennett’s theories on consciousness suggest that intelligence is merely an emergent property of information processing, much like what we observe in modern AI models.

This talk will critically examine these perspectives, discussing whether AI is on the path to achieving human-like cognition or whether there are fundamental barriers preventing computational models from replicating consciousness. Ultimately, we will explore whether the rapid scaling of AI is bringing us closer to AGI or revealing the limits of algorithmic intelligence. Is the human brain super-Turing-powerful?

How to deal with unsuccessful GAČR project applications?

What are the possible reasons for project proposals being unsuccessful? What can we do about it? Come and hear about the experiences of GAČR panel members and discuss ways to increase the success rate of our proposals. We will present our proposals and would welcome your feedback and any suggestions you may have. Given that research funding is a critical aspect of a healthy research environment at the faculty, we must address the existing problems and take the necessary measures. By taking part in this discussion, you can play a direct role in shaping these measures.

Multimodal TimeViz: Visualisation and Multimodal Exploration of Time-Oriented Data

In this talk, I will explore the complexities and nuances of visualising time-oriented data. Time, as a unique dimension, possesses a rich semantic structure that extends beyond a simple linear progression of values. I will demonstrate why effectively modelling this structure is essential for capturing temporal irregularities and cyclic patterns. Through practical examples, I will illustrate how these temporal characteristics significantly impact visualisation design and interpretation. The talk further explores emerging research directions, including visualisation literacy, audio-visual analytics, situated visualisation, and knowledge-assisted visual analytics.

A data-driven approach to medical education and healthcare in practice

This lecture provides a comprehensive overview of the outputs and activities within the academic and governmental sectors that focus on the collection, processing, analysis and visualisation of health and medical education data, including their correct interpretation. It is essential to recognise the importance of correctly managing, identifying and disseminating accurate information and insights from the vast array of data of varying quality.

The lecture includes case studies that illustrate new findings and best practice in technology-enhanced learning and health informatics, drawing on experiences from research and development projects, stakeholders’ views and community needs. This lecture highlights proven methodological and technological foundations that provide a comprehensive and robust framework for two distinct and multidisciplinary areas of human interest and understanding: firstly, medical and healthcare education supported by the MEFANET network; and secondly, health information and statistics, primarily based on data from the National Health Information System. These topics are set against the backdrop of achieved results applied in daily practice, the sharing of know-how, student involvement and community building. Generally, this field of research is multidisciplinary and summarises the results achieved, combining the fields of informatics, medical education, pedagogy, data analysis and health literacy.

What is my quantum programme doing?

Developing quantum programmes is difficult due to their intricate structure and inherently probabilistic nature. Computer-aided tools are therefore essential. However, computer-based reasoning about quantum programmes is also challenging due to the exponential size of the programme’s state. In this talk, I will present a recent framework for the automated formal verification of quantum programmes that uses automata to represent complex sets of quantum states compactly.

CoFI break with management

Lab Fest, held as part of the Computer Science Colloquium, aims to familiarise the academic community with the activities of research groups at the faculty, specifically their staffing, areas of interest, involvement in grants, and ongoing collaborations at the university, in the Czech Republic and internationally; their future direction and current research and development results.

Jan Křetínský: Learning in Verification group (LiVe Lab)

Michal Kozubek: Centre for Biomedical Image Analysis (CBIA)

Jumping from Academia to Startup-Land: One Man’s Story of Transferable Skills

Leaving academia for an academic spin-out can, at first, feel like starting any other research project. But at some point, the differences start to become clear. Whilst the technical challenges may be familiar, the priorities, constraints and measures of success are often quite different. The objectives of a research project do not always align with those of a company; the motivations of funders shift from knowledge generation to commercial impact; and managing a team in a start-up comes with its own unique dynamics.

In this talk, I’ll share my experience of navigating this transition: the lessons I’ve learnt, the mistakes I and others have made along the way, and the aspects of academic training that have proved useful. Whether you’re considering a similar move, want to make your teaching more relevant to students entering the industry, or are simply curious about how research skills translate to start-ups, I hope my talk will provide you with some useful insights.

Kristýna Pekárková: Twin-width of Graphs on Surfaces

Twin-width is a structural graph parameter introduced in 2020 by Bonnet, Kim, Thomassé and Watrigant. Graphs of bounded twin-width generalise a wide range of graph classes, both sparse and dense, and the parameter has proved to be an important tool in many areas, including structural and algorithmic graph theory, combinatorics and model theory. Although it is known that many graph classes have bounded twin-width, the bounds are often enormous, and determining better, explicit bounds has, therefore, been the focus of an intensive line of research. In this talk, we present a new, asymptotically optimal upper bound for the twin-width of graphs embeddable on surfaces.

Ján Jančár: Peeking into the black box: Reverse-engineering cryptographic implementations

Cryptography secures our online communications, payments and other services. We tend to rely on it without questioning its security or that of cryptographic implementations. Software implementations, particularly open-source ones, are auditable, although this is no guarantee of security. However, cryptographic implementations in hardware, such as smartcards, are closed-source, with extremely limited publicly available information, and their details are protected under non-disclosure agreements (NDAs). Users therefore have no way of independently verifying the security claims and are left to trust security certifications that have been shown to be flawed in the past.

In this talk, we will discuss several methods of ‘peeking into the black box’ and gaining insight into the implementation details of elliptic curve cryptography. We will see how simple algebraic tricks can be used to extract closely guarded details from smartcards produced by major manufacturers. Finally, we will discuss what this means for the secure hardware sector and what manufacturers can do to protect themselves.

What challenges do biomolecules pose for IT?

Biomolecules, such as proteins and nucleic acids, are 3D objects that form the essential building blocks of living organisms. But what do they contain? Various complex data structures – sets of 2D and 3D points, graphs, multigraphs, matrices, groups, etc. The contents of these data structures make sense not only biologically but also logically. Moreover, the datasets relating to biomolecular 3D structures are vast. Furthermore, research into these data can yield breakthrough results that improve our lives.

In my talk, I will present selected key challenges in biomolecular 3D structure research (e.g., structure visualisation, prediction, similarity search, property computation) and IT approaches to addressing them. In addition, I will highlight the questions that remain open and await IT solutions.

FIxbot: Practical steps towards AI-assisted teaching at our faculty, discussion

The talk will outline the initial demand, visions and practical experience with AI-assisted teaching at our faculty, highlighting the need for changes to teaching methodology as well as self-hosted models with customisable processing pipelines to provide better control over interaction data for improved education.

The rise of LLM-powered programming assistants has upended conventional teaching strategies, particularly in programming courses where step-by-step skill-building has been the norm. Whilst these models excel at solving simpler problems, they leave students struggling when faced with more complex assignments — without the foundational understanding built through progressive learning.

Our approach is based on the assumption that students – future graduates – will routinely use LLM-powered programming assistants in their work. Therefore, the learning outcomes of programming courses should focus on collaborative development, in which the programmer interacts with the assistant. The developer’s responsibility is to take ownership of the result, which involves thorough verification of compliance with the specification, testing and debugging.

We have designed FIxbot – an LLM-based pipeline that inserts bugs into the code and provides guidance on identifying these bugs when asked. We will summarise observations and lessons learnt from using such a pipeline to teach PB071 (Programming in C), including the technical deployment of open-weight models within the FI infrastructure. Additionally, we will explore how students on PB173 (Domain-Specific Development) utilise a range of commercial and open-source LLM-based setups for their programming tasks.

The development of FIxbot.fi.muni.cz is a work in progress, and we need your insights and expertise. A significant portion of the talk will be devoted to discussions about further uses of FIxbot at FI, new ideas and technical considerations.