Mathematical Foundations of Artificial Intelligence

Artificial intelligence (AI) has long been one of the big promises of computer science, and related research has fundamentally shaped our thinking about the human-machine relationship. For decades, AI has failed to fulfil its promise of solving practical challenges. This is changing now. As the past ten years have given rise to radical changes and major breakthroughs making AI the major solution in many situations.

Machine learning is a set of mathematical techniques which powers most current-generation artificial intelligence systems, notably self-driving cars, intelligent assistants, speech recognition and machine translation software. Large corporations have put a vast amount of effort into improving these techniques, but the theoretical underpinnings of why and how they work are still undeveloped.

The goal of our proposal is twofold: first, we wish to bridge the gap between mathematical theory and machine learning practice. More specifically, we wish to exploit newly discovered deep connections between fundamental results related to the study of large networks — an area to which Hungarian scientists have contributed tremendously — and the more applied domain of machine learning. Second, we wish to firmly establish machine learning as a line of research in Hungary, where applied machine learning is underrepresented.

We aim to achieve our goals by establishing a competence centre which integrates theoretical and applied research, providing the background for domestic practical applications. Crucial roles of this initiative are, on one hand, to connect the country into the network of international research, and on the other, to facilitate machine learning applications in the industry.

Beyond fundamental research, we intend to carry out an interdisciplinary pilot project which directly demonstrates the practical applicability of our research. The aim of this project is to apply machine learning to enhance the treatment of chronic wounds, in close cooperation with medical practitioners. Today, around 200 thousand people in the country are suffering from chronic wounds. Increasing their quality of life as well as reducing costs by finding the optimal treatment would have a huge societal impact. Furthermore, the success of this project would set a precedent for later interdisciplinary machine learning research.