'ISSCAI Greatly Immersed in IT Environment': HSE University-St Petersburg Finishes School and Conference on AI
HSE University-St Petersburg held the International Summer School & Conference on Artificial Intelligence—ISSCAI 2024. It was organised by the School of Physics, Mathematics, and Computer Science together with the Laboratory for Social and Cognitive Informatics supported by VK and other partners. A one-week intensive course gathered talented students, PhD students and developers from all over the country—from Moscow to Yekaterinburg as well as Kazakhstan and Egypt. Especially for them, leading experts from Russia, the UAE and India delivered courses on key trends in the sphere of artificial intelligence: from computer vision to music signal processing.
The technologies of artificial intelligence are regularly developing, and HSE University-St Petersburg is exploring different facets of its use—for instance, in education or media. Modern business cannot do without AI: recent research by the NAFI Research Centre shows that every third representative of small and medium-sized businesses in Russia uses it in their work. Therefore, it is becoming increasingly relevant to prepare IT specialists who have skills in working with various AI technologies. To introduce students and novice developers to the latest trends in this sphere, the School of Physics, Mathematics, and Computer Science in cooperation with the Laboratory for Social and Cognitive Informatics held the International Summer School & Conference on Artificial Intelligence (ISSCAI) on July 1-7.
For the participation, based on resumes and summaries of research theses, the organisers selected 50 motivated and qualified students, PhD students and young developers from the leading Russian universities: MSU, TSU, BMSTU, SPbPU, VSU and others. During a week, they attended five courses and four guest lectures by 'Technologies of Trust', 'Sber', VTB, Synthesis AI, VK, MSU and the Speech Technology Centre.
The general partner of the project was one of the IT industry leaders—VK company. VK offered the school attendees to discuss all work aspects of the specialists in ML and DS in industrial development—from setting tasks to maintaining the finished service. The students listened to the presentation by Vladislav Efimov, Tech lead of E-commerce team at VK. The industrial partners of the school and conference were also the company 'Technologies of Trust' (TeDo) and VTB Bank.
Apart from the Russian community of IT specialists, ISSCAI gathered foreign ML researchers. Thus, HSE University-St Petersburg hosted for the first time a professor from the UAE—Baris Bozkurt, Associate Professor at Zayed University, Abu Dhabi. He devoted his course to difficult questions: how to analyse and represent musical signals with the help of classical methods of machine and deep learning.
Baris Bozkurt, Associate Professor at Zayed University in Abu Dhabi
In Turkey and the UAE, a common opinion about Russian IT specialists is that they are well-educated and have strong mathematical and theoretical backgrounds. During the school and conference, I managed to see it for myself: there are energetic and interesting students and amazing professors. I hope we will continue our cooperation.
Computational musicology is a vivid interdisciplinary area. One of the key and perhaps the most difficult questions in this sphere is the development of automatic systems of data mark-up and classification as manual mark-up is becoming a more difficult task. For instance, during musical education, it is required to provide feedback on the students' performances. We have some tools which are used in the software for musical education but these are only the first steps. This research field still has a lot of open topics which makes it attractive for many specialists.
Professor Mausam from India also shared his expertise with the students. He offered the attendees to immerse themselves in difficult NLP topics such as aligning of language models, generations of answers based on relevant information and reasoning with language models. Professor Mausam represented the Indian Institutes of Technology in Delhi which has been a partner of HSE University-St Petersburg in the sphere of social, political and humanitarian research and data analysis since September of this year. The Institute is one of the best not only in India but in the whole world: in QS World University Rankings, IIT Delhi takes the 150th position.
Olessia Koltsova, Head of the Laboratory for Social and Cognitive Informatics
Our idea is to create at HSE University-St Petersburg a hub uniting the international AI community of different clusters: industry and academy representatives and young talents—students and developers. The students get to know innovations in the AI sphere, IT companies get an opportunity to involve novice specialists, and HSE University-St Petersburg increases its visibility in the sphere of machine learning both in Russia and abroad. In their turn, attendees and professors who came to us from other countries got to know the Russian IT industry. We believe that our main achievement is that we managed to attract very interesting high-level speakers, including international ones from the leading universities in India and the UAE. We hope that we will manage to make such events regular.
ISSCAI went in close cooperation with business. Thanks to lectures and courses by practitioners in the sphere of machine learning, the conference participants managed to learn how Russian companies apply artificial intelligence to solve business tasks. For instance, during the three-day course by TeDo, the students found out the advantages of popular architecture for processing an artificial language—'Transformer' and the relevant issues it helps to solve. In its turn, VTB Bank shared the key results of the four-year transformation of the company's IT landscape. One of them is the creation of the platform for implementing AI models which allows the employees to create and deploy AI models at the industrial level.
Denis Surzhko, Deputy Head of the Department of Data Analysis and Modeling, Vice-President, VTB Bank
The technologies of artificial intelligence rapidly developed in recent years are completely changing the approach to running a business in several spheres. Over four years of work, VTB Bank has undergone a digital transformation and received a new IT landscape based on the fundamental data processing architectures. The key role in it is given to the platforms for managing and implementing IT models which help to monitor regularly, train and introduce models against the background of high-speed changes in science, economics and the social sphere.
A peculiarity of the bank is the maximum requirements for safety and data confidentiality. In this case, the help comes from the synergy of AI technologies and data security—crypto enclave. It helps to control the quality of data and develop AI models without privacy violation by using an approach of 'blind' machine learning.
The courses and lectures during ISSCAI highlighted that AI technologies gradually enter all spheres of people's lives. However, this fact motivates the researchers to think about the following issue: how soon AGI (artificial general intelligence) will appear—a truly strong artificial intelligence which will be equal to human’s or even excel? This issue was covered by the lecture by Sergey Nikolenko, Head of AI at Synthesis.
Sergey Nikolenko, Head of AI at Synthesis
The spring of last year can be rightfully called the spring of artificial intelligence: almost every day, we learn about new interesting results in machine learning. March 14 alone was masked by three events which could become a theme to discuss for several months even a couple of years ago: for instance, OpenAI presented GPT-4, and Google integrated LLM in the Workspace system. Users are mastering Chat-GPT and prompt engineering quite quickly. This area is developing so rapidly that researchers believe that we are on the verge of creating artificial general intelligence—AGI—a model equal to a human brain. In fact, it is not so important if it will be able to do what people do. What is more important is if it will be able to conduct research in the AI sphere, which means to improve itself. But when will it happen and what consequences will it bring to people? Of course, these are questions without answers but they worry lots of scientists. For instance, Geoffrey Hinton, Yoshua Bengio, Sam Altman and other founding fathers of AI signed an open letter calling to make the reduction of potential risks, possibly caused by this technology, a global priority.
Such events like ISSCAI make HSE University-St Petersburg an attraction centre for leading researchers and interns in the sphere of artificial intelligence. However, HSE University-St Petersburg itself is also actively developing this technology. Recently, the School of Physics, Mathematics, and Computer Science won the HSE Contest of Competencies. Thanks to this success, the school and its partners will create an educational AI assistant for students. What is meant here is a recommendation system allowing one to build an individual educational trajectory taking into account the skills demanded in the labour market.
The project will also feature the Laboratory for Social and Cognitive Informatics of HSE University-St Petersburg, SberDevices, SPCRAS, SPbPU and commercial companies. In the final stage, Sergey Nikolenko will join the project team as a consultant. Trying their hand at creating real-life solutions based on AI will be also available for students of the Bachelor's programmes 'Applied Data Analysis and Artificial Intelligence' and 'Applied Mathematics and Information Science' as well as master's programmes of the School.
A one-week school and conference finished with a poster session where students presented their researches to IT specialists. The participants shared that ISSCAI helped them to immerse themselves in the latest innovations in the sphere of AI and establish useful contacts from all over the country and abroad. In their turn, the partners and IT companies highlighted that they planned to continue the cooperation with HSE University-St Petersburg.
Zoya Minyushkina, Plekhanov Russian University of Economics
I am a product manager, and it was interesting for me to look at the IT industry from the inside. ISSCAI gave me such an opportunity. The communication with amazing professors and experienced students was very productive and inspired me to new ideas in my sphere. I especially liked the lectures by Professor Baris Bozkurt—it was unexpected and truly interesting that you can approach the study of music from the point of view of mathematics and new technologies.
Vladislav Kalinichenko, Innopolis University
I truly liked how this school and the communication with the participants were organised. Thanks to it, I gained amazing contacts with other students and found like-minded people from all over the country—it is wonderful and useful. Of course, I was delighted by the expert lecturers. The course on Advanced Natural Language Processing by Professor Mausam was the coolest of all I've ever attended. Professor Mausam spoke so clearly and fascinatingly about transformer neural networks that for the first time, this topic didn't seem so difficult for me. Besides, I had not expected to meet here one of the most quoted data scientists in Russia—Sergey Nikolenko. Now, I tell all the coursemates that I was sitting next to him and listening to his lecture. To sum up, ISSCAI immersed us in the IT environment greatly.