Automated discovery systems appeared in the 70s; full-fledged experiment designs using decision theory and causal models were reported around 2000, and subsequently, their robotic extensions with planning soon emerged. In the last decade, large-scale quantitative studies analyzed policies for the selection of scientific experiments and the evolution of science. Artificial intelligence (AI)...

6. Building quantum computers - engineering challenges

Anna Kaminska (Creotech Instruments S.A.)

The current status of trapped ions quantum computer development in Europe will be presented from the industry perspective, with focus on main challenges for the next few years. A brief overview of practical aspects related to scaling the number of qubits and improving the performance of quantum computers will be given. Special attention will be devoted to engineering challenges, in particular...

9. Classifying ECG signlas using linear laws

Peter Posfay (Eötvös Lóránd University)

I introduce a new method called linear law-based feature space transformation (LLT) which can be applied in the analysis of time series. This method builds on ideas from physics and data science. It implements a transformation of the input time series in such a way that the resulting feature set can be effectively used for classification tasks. After describing the method, I present its...

3. Natural Language Understanding using the Concept Network Model

Zoltán Foris (Human Mind Project)

While deep learning based systems have recently produced significant advances in their capability of authoring sensible text, some researchers argue that a different approach is needed to achieve true language understanding. I will present the Concept Network Model of human thinking and comprehension, which is a hybrid symbolist-connectionist approach. This model draws its motivation not from...

12. No-signaling and quantum correlations from the applications' point of view

Mátyás Koniorczyk (Wigner FK)

Bell's inequalities were the first quantifications of quantum nonlocality. They are relevant in the so-called Bell-type scenarios and appear as a consequence of quantum entanglement, and the Nobel prize in Physics in 2022 was awarded partly for their experimental study. This kind of nonlocality is the key resource of the emerging technology of quantum information processing, including both...

2. On the Role of Neural Collapse in Transfer and Few-Shot Learning

András György (Deepmind)

Using representations learned by large, pretrained models, also called foundation models, in new tasks with fewer data has been successful in a wide range of machine learning problems. In particular, recent results in the literature show that representations learned by a single classifier over many classes are competitive on few-shot learning problems with representations learned by...

11. One Lab Many Projects -- Review of the Wigner Scientific Computing Laboratory -

Gergely Barnafoldi (Wigner RCP RMI of the Hungarian Academy of Sciences)

The Wigner Scientific Computing Laboratory (WSCLAB) has been founded in 2021 December on the basis of the Wigner GPU Laboratory and the WLCG ALICE/CMS T2 site. Here I plan to give a short color scope on the projects for the last 12 years.

1. Preventing and restoring adversarial attacks in machine learning solutions

Andras Horvath (Pázmány Péter Catholic University - Faculty of Information Technology and Bionics)

Modern machine learning enabled the solution of complex problems, where a vast amount of data is available.
Unfortunately in many cases the developed solutions are not robust and even the slightest modifications can drastically change the networks output. The most interesting deliberate modifications are the so called adversarial attacks, which can be considered as optical illusions for...

13. Quantum Key Distribution: theory and practice

Tamas Kiss

Measurement in quantum mechanics is of probabilistic nature. Furthermore, in a given state of a quantum system, it is impossible to measure the value of two (noncommuting) observables without disturbing the state. These features of quantum mechanics provide a way of generating a sequence of correlated random numbers at distant places, where any disturbance by an unfriendly agent can be...

5. Spiking Variable Projection Neural Networks for Arrhythmia Detection

Péter Kovács (Department of Numerical Analysis, Faculty of Informatics, Eötvös Loránd University)

Edge-based Machine Learning (ML) has a pivotal role in revolutionizing smart healthcare by introducing a tangible improvement in the secure and discrete medical data analysis.
This paper presents a novel neural network architecture by combining Variable Projections (VP) and Spiking Neural Networks (SNN). VPs are nonlinearly parameterized orthogonal projections whose weights have physical...

7. System identification with rational orthogonal functions

Mátyás Szabari (ELTE TTK)

System identification is a data-driven approach to model the behavior of dynamical systems. Most algorithms of this type have a black-box view of the system, i.e. they relies exclusively on the observed input/output measurements, rather than utilizing the underlying physics. The identification process can be done in both time and frequency domains having various trade-offs, such as...

8. Testing Rankings with Cross-Validation

Balázs R. Sziklai (Centre for Economic and Regional Studies (KRTK))

Ranking objects is one of the most commonly applied computational tasks. In machine learning, query results are ranked by search engines, features are ranked during feature selection, algorithms are ranked by their performance. Often there is a reference through which the solutions are compared to each other. In content recommendation, this can be the user for which we would like to generate...

10. UAVs collaborate with satellites in an early wildfire monitoring and rescue system

Chang Liu (Machine Perception Research Laboratory of Institute for Computer Science and Control (SZTAKI))

Wildfires pose a significant threat to human safety and animal mortality, and it is becoming increasingly important to monitor and stop the onset and spread of these wildfires. Drones have the advantages of small size and flexibility, making them one of the best tools for monitoring these early wildfires. With the rapid development of drone technology, it has become possible to use UAVs with...