Research Topics
Theoretical Computer Science: Algorithms for dynamic graphs. Parametrized complexity. Fixed points in computer science. Automata and formal logic. Lexicographic orderings of languages. Tree automata and tree languages, tree transducers. Term rewriting systems. Automata and semirings, formal power series. Weighted tree automata. Grammar systems, formal language models of distributed and cooperating systems. DNA computing, molecular computing. Membrane systems, computational complexity of active membranes.
Operations Research and Combinatorial Optimization: Theory of economic decision making (multicriteria decisions, team decisions). Fuzzy theory. Learning algorithms. Global optimization. Reliable numerical procedures. Interval inclusion functions. Process network synthesis. Bin packing algorithms. Online algorithms. Scheduling problems. Game theory. Facility location. Transportation problems.
Software Development: Artificial Intelligence for Software Engineering (AI4SE). Software Engineering for Artificial Intelligence (SE4AI). Traditional and artificial intelligence-based methods for program analysis, software testing and software maintenance. Databases, data mining, data science. Theory of compilers. Efficient programming of embedded and mobile systems. Program slicing and its applications. Network protocol analysis and testing. Parallel programming. Studying IoT, Cloud and Fog systems. Analysis of blockchain systems.
Artificial Intelligence: Machine learning algorithms (decision trees, genetic algorithms, neural networks, deep neural networks). Complexity of learning algorithms. Speech recognition, speech technology. Language technology, natural language processing. Semantic Representations. Interpretability. Human-machine interface, dialog systems. Distributed machine learning. AI Security, robustness. Medical applications of AI.
Image Processing: Image reconstruction from projections. Discrete tomography. Medical image analysis. Image segmentation. Image registration and fusion. Computer vision. Skeletonization, thinning and their applications. Discrete geometry and topology. Statistical image models. Markov random fields. Deep Neural Networks. Camera pose estimation. Localization and navigation. 3D reconstruction. Point cloud processing. Object detection. Remote sensing. Microscopic image processing. Motion detection and tracking. Variational and level set methods. Industrial image processing. Non-destructive testing.
Technical informatics: FPGA-based image and signal processing. Sensors, sensor networks, embedded systems, sensor-based signal processing. Noise and fluctuations in different systems, noise analysis of movement patterns, noise-based secure communication. Software-defined instrumentation, measurement and processing of physiological signals. Development of modern tools and methods for STEM education.
Process
There are two ways of obtaining the PhD degree: by following a four-year study program, or by individual preparation. The program of which the duration is 4 years prescribes the accomplishment of 240 credits, active participation in the Institute's seminars, and the conduction of research under the supervision of a thesis adviser appointed by the Council of the Doctoral School. At the end of the fourth semester, the completion of 5 courses is included, the courses embrace a number of fields in computer science without the intention of being exhaustive. A course may be offered as a reading course if enrollment is low. In such cases consultation is provided. The language of education in the four-year program is mainly Hungarian, but for foreign students, each course may be offered in English. The teaching and research staff of the Doctoral School consists mainly of scientists working at the Institute of Informatics and the Research Group on Artificial Intelligence of the Hungarian Academy of Sciences. Some members of the Institute of Mathematics (Faculty of Sciences and Informatics), Department of Medical Physics and Informatics (Faculty of Medicine), and the Department of Applied Informatics (Juhász Gyula Teacher Training College Faculty) also participate in the School. Foreign lecturers may also announce courses in the Doctoral School. The requirements for obtaining the PhD degree are the following. At the end of the second year, each candidate has to pass a comprehensive doctoral exam which has two main parts. During the first (theoretical) part, the student takes exams in one major subject/topic and in one minor subject/topic. In the second (dissertation) part of the comprehensive exam the student holds a lecture, giving account of his/her knowledge about relevant scientific literature and his/her research results, and describing his/her research plan for the second part of the doctoral training, and the schedule for writing the dissertation and publishing the results. Regarding language skills, a state examination or equivalent at a level not lower than intermediate is required in a foreign language accepted by the School. A lower level examination or equivalent is required in a second foreign language. One of the two languages must be English. Special rules apply to foreign students. As a third major requirement, each applicant has to fulfil the publication requirements of the Doctoral School. As a last major requirement, each candidate has to prepare and defend a doctoral thesis containing new scientific results in computer science, or high-level applications of computer science in other areas. A dominant part of the results included in the thesis has to be published before submission.