Scientific paper at the 23rd INFOTEH-JAHORINA conference

A scientific paper “Output Manipulation via LoRA for Generative AI” by I. Culafic et al., was presented at the 23rd International Symposium INFOTEH-JAHORINA, 20-22 March 2024. The training for the prediction models was takin around six hours on an NVIDIA RTX 4090 24GB VRAM GPU. This research will serve as a basis for a future experiments on HPC resources. The paper is published at IEEE Xplore at: https://ieeexplore.ieee.org/document/10495995

ABSTRACT – Generative Artificial Intelligence has witnessed a surge in popularity in recent years, characterized by the emergence of groundbreaking models like DALL-E 2,
Midjourney, and Stable Diffusion, which have spearheaded advancements in this technological domain. This research aims to harness the potential of Stable Diffusion and its extensions for the purpose of training a LoRA (Low-Rank Adaptation) model to
generate images that closely resemble the original subject matter, utilizing a predetermined amount of example data. The primary objective of this research is to demonstrate the prowess of Stable Diffusion and generative AI in a broader context, delving into the possibilities offered by open-source frameworks, highlighting the
challenges associated with poorly organized training data and the advantages of properly organized and edited datasets, conducting a comparative analysis of diverse diffusion models and examining various LoRA strength examples. This research also aims to
compare the results from larger training parameters on both small and relatively large training models for the purpose of determining if overfitting, over training on one specific subject, is more prevalent with smaller or larger datasets.

HPC and AI Workshop

NCC Montenegro and NCC Spain, collaborators at the EuroCC project, organized the “HPC and AI Workshop” designed for academia, scholars, researchers who are interested in learning how to use Artificial Intelligence and High-performance Computing in real case scenarios. This has been achieved through five lectures featuring educators from Spain and Montenegro presenting successful case studies:

Borja Perez Pavon, from Universidad de Cantabria, talked about HPC systems and highlighted they are available for different research projects by accessing various free access calls.

Ivan Jovovic, AI researcher and PhD candidate at the University of Donja Gorica explained from the use of multiple GPUs to handle large data sets to the cons and pros of data parallelism.

Stevan Cakic is a PhD student at the University of Donja Gorica, talked about the application of AI and HPC in agriculture, delving into deep learning and high-performance computing.

Ahmad Al Mughrabi, active researcher and multidisciplinary predoctoral scholar at the Universitat de Barcelona, discussed the applications of HPC and generative AI, acknowledging certain challenges that need consideration in the implementation of HPC in real-case scenarios.

Dejan Babic, a PhD candidate at the University of Donja Gorica presented the concept AI and HPC use cases in medicine.

The 28th international conference of Information Technology 2024 (IEEE IT2024)

In the period from 21st – 24th of February the international scientific and professional conference “INFORMATION TECHNOLOGIES 2024” will traditionally be hosted in Žabljak. These 28th years in a row scientific and professional conference is organized with the aim of a comprehensive and multidisciplinary view of current and development trends in the field of information and communication technologies.

The conference will be held in the organization of the University of Montenegro – Faculty of Electrical Engineering, University of Donja Gorica – Faculty of Information Systems and Technologies, IT Society Montenegro, University of Belgrade – Faculty of Organizational Sciences, Institute of Electrical and Electronics Engineers – IEEE Association and IEEE Section for Serbia and Montenegro, with full support of the company Čikom from Podgorica.

Click to watch the video announcement (in Montenegrin)

The Conference shall host lectures and round table discussions about development trends in the field of information and communication technologies, as well as actual problems in this field in Montenegro. In agreement with the Organizing Committee of the Conference, interested institutions are invited to organize presentations of their scientific, research, professional, development, and production projects and achievements. Besides mentioned above, papers submitted and reviewed will be presented at the Conference.

We invite all those interested to follow the activities at the Conference online as well, through the video conference access links available on the official website

Learn more at the conference website: https://www.it.ac.me/en/

Click on image to open IT2024 conference website

Online event: Artificial Intelligence and High-performance Computing for real case scenarios

SAVE THE DATE: January 29th, 11:00 AM, register for free: https://bit.ly/3tGjim3

We are organizing an online training event on Artificial Intelligence and High-performance Computing for real case scenarios. This workshop is designed for academia; scholars, researchers, students, who are interested in learning how to use AI/ML and High-performance Computing in real case scenarios. This will be achieved through five lectures featuring educators from Spain and Montenegro who will present successful case studies. This is a twinning effort by two NCCs.

Agenda for the online training by NCC Montenegro and NCC Spain

Hackaton on the use of Computer vision in Montenegrin S3

During the 3 months of intensive work, NCC Montenegro team members were involved in organising 70 classes for more then 50 participants. Lectures were held twice a week, starting in October this year. The training started with Python programming module which focused on mastering all programming skills needed for further work and engagement with Artificial Intelligence and Machine Learning. Then we explored AI concepts, mathematical models, computer vision tasks, and provided hands-on examples of using Convolutional Neural Networks, Yolo architectures (YoloV7 and YoloV8), and the Detectron2 package for various computer vision applications. Throughout these modules, participants not only gained exposure to various tools, Python packages and libraries for facilitating AI model development but also had the chance to learn the integration of AI solutions into modern systems and applications with Flask and other web bases technologies. The attendees were informed on NCC Montenegro activities and introduced to capabilites of HPC and its application for AI development that requires lots of computing resources.

Hackaton took place on 23 Dec 2023.

After all classes we organised Hackathon day. The Hackathon gathered 20 teams to work on HPC/AI and Computer Vision projects in agriculture, medicine and tourism. Before the Hackathon Day, after which the best ideas were showcased and awarded, there were two weeks of intense mentor-guided sessions. These sessions helped teams with their projects and in figuring out high-performance computing needs for handling large datasets , execute experiments, and testing advanced AI models. This hackaton was implemented in cross-collaboration with the AI4S3 project funded by the Innnovation Fund of Montenegro.

The hackaton was attended by over 50 people
24 teams presented their potential application of HPC and AI in Montenegrin S3
The presentations focused on health, tourism, energy and food sectors
The hackaton ended in celebration and social event

Workshop “Digital Transformation and HPC/AI” for BSc students

On 8 Dec 2023, a workshop on “Digital Transformation and HPC/AI” unfolded at the University of Donja Gorica. Over the course of this semester, final-year BSc students from the Faculty for Information Systems and Technologies delved into research on HPC and AI technologies within the framework of the Managing Information Technology subject. The EUROCC NCC Montenegro team conducted enlightening lectures and presentations on HPC and AI, providing students with a comprehensive grasp of these technologies and illustrating their pivotal roles in digital transformation.

Digital transformation and HPC/AI students workshop

During the workshop, students had the opportunity to showcase the projects they worked on throughout the semester, projects that will be defended during their final exam in Managing Information Technology course. The presented use cases spanned across various fields such as tourism, medicine, agriculture, and digital marketing. This event not only allowed students to share their findings but also fostered a deeper understanding of the practical applications of HPC and AI in real-world scenarios. Our focus was on the possible applications of interest for Montenegro and in the priority domains of Smart Specialisation Strategy for Montenegro (2014-2019).

Over 20 BSc students participated
AI powered chatbot for hotel reception
HPC/AI model cretion for sign language
HPC/AI for recognizing cultural and historic monuments
Prediction of attendance of sport events
Use of computer vision in medicine
Computer vision for drones for prevention of forest fire
Computer vision for precision agriculture and pest management
Event took place at the University of Donja Gorica

Progress on AI4S3 project

In the previous period, as part of the AI4S3 project, which is supported by Innovation fund of Montenegro, numerous lectures on the application of artificial intelligence in S3 areas were organized at the UDG:

  • Python programming: Our participants went through a thorough training in Python, starting with the basics of the programming language, all the way to advanced topics. Through homework, our participants demonstrated a high understanding and application of Python in various scenarios. This was the first step towards strengthening their programming expertise.
  • Introduction to Artificial Intelligence (AI): this module was a journey through the world of artificial intelligence. We started with the historical development of the technology, exploring how machine learning and deep learning algorithms have evolved over time. In the pyTorch and TensorFlow suite, we developed some basic deep learning models. We focused especially on neural networks, explaining their mathematical and technical foundations. Participants gained a thorough understanding of the technologies that shape our world.
  • AI in S3 priority areas: this segment was extremely important because the basic idea of this training is precisely the application of new technologies. Innovative projects developed by our lecturers, specifically focused on the application of AI in S3 areas, were presented. This gave participants an insight into the practical application of AI in the real world. Also, we dealt with ethical issues and challenges in the development of AI models, encouraging participants to think critically about the impact of technology on society.
The training included Python basis, introduction to AI and ML, applications in S3 domains

With over 50 participants from various academic and industrial circles, our training has become a meeting place for different perspectives and innovations. By the end of this training, two more modules are planned, namely The Role of Artificial Intelligence in Computer Vision and Tools and Libraries for AI Model Development. At the very end, training participants will present their projects, which they will develop in teams in cooperation with mentors. This training is a continuation of previous NCC Montenegro and Open Mind Academy efforts.

Mr Stevan Cakic, AI4S3 project coordinator
Over 50 trainees from various academic and industrial sector
Participant will be trained and metored to implement their own hands on projects
The course included cross-collaboration and presentations of practical use cases