Artificial intelligence (AI) algorithms are typically used to solve the most challenging problems, achieving state-of-the-art performances and providing a substantial impact on all aspects of human life. However, because of their computational complexity and high number of parallel processing operations, AI algorithms require significant computational resources (processing power and memory), they consume a considerable amount of energy and require powerful hardware (such as GPUs, servers, clouds). All of these constraints, significantly limit their applicability. In recent years, there has been an increasing need for implementation of AI algorithms on devices with limited resources (memory space, processing power and available energy). Therefore, the problem of implementing complex AI algorithms on devices with limited resources is one of the most topical and challenging research problems in the field of AI. This project deals with solving the mentioned problem, where the overall objective is to propose and implement new approaches to reduce the complexity of AI algorithms using advanced quantization and compression methods, based on the strong expertise of the project team in the field. An important research question we will answer is how to make state-of-the-art DNN, with numerous parameters, more compact and efficient. We will exploit the fact that reduced precision of the DNN parameters, i.e. going from 32-bit floating point to a low-bit fixed point, can be achieved by thoughtful application of quantization and compression, which will result in a worthy compression ratio and, accordingly in the reduced storage and energy cost, and computational requirements. Majority of the research directions has been set toward achieving the highest possible compression ratio of DNN parameters without significant accuracy loss. Since this field of research is still in an early stage, significant improvements are possible. This project offers new approaches to advanced quantization and compression of DNN parameters in order to outperform state-of-the-art solutions. Besides their fundamental scientific importance in the field of AI, all theoretical results will also be fully applicable in practice. Hence, an important part of the project concept is to validate all the proposed theoretical results and to examine their applicability. This will be done by implementing all theoretical solutions on the developed DNN, as a part of the target use case. Eventually, all theoretical results will be implemented and validated and their performance will be evaluated within a real industrial application using data from a real industrial environment.
Very topical issues present in generally very powerful AI algorithms related to decreasing computational complexity and memory resources that are of particular importance in portable and edge computing devices with limited memory and processing power are the driving force behind this project proposal. Specifically, this project contributes to a new and fastly growing research in the worldwide AI science solving these particular issues by prudent and deliberate application of quantization theory. Due to the topicality of these issues and the fact that related research is still in the early stage and deserves further investigation, our project team will explore and propose innovative methods of compression and quantization of DNN parameters (weights, biases, activations) and deep features. Also, the goal of the project is to develop a state-of-the-art deep neural network model with a high performance not only on the hardware usually used for AI applications but also on devices with limited computational resources and thus enabling them to support energy demanding and memory constraint applications. In order to achieve these goals, the project proposes an integrated approach to quantization and compression of DNN parameters, based on statistical modeling of the data per layers, as well as of the data subsets within layers and adaptation of the quantizers itself on the statistical characteristics of the input data. Moreover, from the exploration of the compression and quantization effects in DNNs vis-à-vis its accuracy, the benchmark of our methodology will be defined. The researchers’ interdisciplinary competence ensures successful development of innovative quantization, compression and learning methodologies that will enable reducing the complexity of AI algorithms and its much wider usage. The results obtained within this project will find a wide range of applications in both academia and industry, particularly in numerous latency-critical services.
The project team is fully compatible, since all team members have basic knowledge of quantization and coding theory, as well as of neural networks and AI algorithms, which will greatly facilitate their collaboration and communication and will improve work efficiency.
Moreover, the project team has a great complementarity since each team member has a unique deep knowledge and strong expertise required for the project realization.
Team members have been working successfully at the University of Niš, Faculty of Electronic Engineering, on joint scientific projects and papers for years and therefore know each other’s strengths and possibilities very well, which will give the optimal outputs of this project.
Professor Zoran Perić as a PI has a stunning academic and scientific career, as well as necessary management and administrative skills to lead this project. With years of experience and significant scientific and applicable results, a Full Professor Dejan Ćirić will give great contribution in the activities within this project related to AI (DNN design, application and evaluation), audio signal processing and sound acquisition.
The special expertise of Associate Professor Aleksandra Jovanović is design of vector quantizers and also design of power efficient signal constellations. Therefore, it is expected that she will predominantly contribute to the part of the project related to data clustering and classification where the multidimensional space partition is an important issue. Assistant Professor Milan Dinčić has strong expertise in the design of VLC (variable-length coding) compression algorithms as well as in the statistical modeling of input data for DNNs and other AI algorithms. His contribution mostly will be in the domain of compression.
Associate Professor Jelena Nikolić will contribute to this project with her expertise in speech compression and its application in machine learning and AI. In particular, she will give great contribution in the project activities related to quantization and compression of DNN parameters, where her expertise can be exploited greatly.
Ph.D. student Nikola Vučić as a young researcher brings fresh energy to the project and provides support to the part related to quantization and compression.
Ph.D. student Bojan Denić will contribute to the part of the project related to data classification, as it is closely related to his research areas.
Professor Vladimir Despotović is an expert in machine learning, natural language processing and fractional calculus. His knowledge and extensive international experience considerably increase the capacity of the team to develop advanced methods of learning in AI. All project objectives will be realized through an interdisciplinary approach and knowledge synergy of all team members. Eventually, team members are carefully selected so that everyone has a precisely defined role in the project.
Principal Investigator
zoran.peric@elfak.ni.ac.rsMembers of our project team have participated in several projects closely related to the topic of this project. PI, P2 and P4 have been involved in the project Human-Machine Speech Communication, related to a detailed analysis of all problems in speech to machine communication in both directions with the aim to give machines intelligence and human-like capability to speak and understand speech. Also, PI, P2 and P4 have been involved in the project The Development of Dialogue Systems for Serbian and Other South Slavic Languages considered the development of dialogue systems using artificial neural networks, which is highly correlated with this project. PI, P5 and P6 have taken part in the bilateral project, Fractional calculus approach to machine learning, between Serbia and Slovakia. The project developed a novel approach to machine learning by introducing fractional-order calculus to optimization methods used in machine learning algorithms. P1 was a project leader of recently finished project SONO360 - Smart audio interface financed by the Innovation Fund of the Republic of Serbia. The project dealt with the development of a prototype of smart audio device capable of acquiring sound from 3D space applying direction of arrival (DoA) and beamforming technique using spherical microphone array, as well as detecting and recognizing domestic sound events based on a specific design of a DNN. There is a close relation to this project with regards to DNN design and implementation in the field of sound. The experience and knowledge gained working on those projects are very valuable for the accomplishment of the tasks and goals we have defined in this project.
November 2022
On November 22, 2022 at the Faculty of Electronic Engineering, University of Niš, Final project promotion was organised, results were summarised and the possibilities of further cooperation were discussed.
November 2022
Prof. Dr. Zoran H. Perić, Dr. Jelena Nikolić, Prof. Dr. Marko Petković and Prof. Dr. Vlado Delić are Editors of the Special Issue "Artificial Intelligence and Mathematical Methods" in MDPI Mathematics Journal.
>>> Link to the Journal <<<
Deadline for manuscript submissions is 28 February 2023.
July 2022
Com-in-AI team members visited the University of Potsdam and the Leibniz Institute for High Performance Microelectronics in Frankfurt (IHP) at Oder. During the visit, closer cooperation was established with colleagues from Germany and it was agreed to jointly apply for future projects.
June 2022
Radio Television of Serbia (RTS) visited our team and faculty for the filming of Science in Motion: Development of Artificial Intelligence through Projects.The film shown on the RTS is also available on the YouTube channel https://www.youtube.com/watch?v=drJhUo8KXvE?t=469
June 2022
The team members Dejan Ćirić, Jelena Nikolić and Nikola Vučić took part in the 57th International Conference on Information, Communication and Energy Systems and Technologies, ICEST 2022, Ohrid, North Macedonia, June 16-18, 2022. The team member Dejan Ćirić also took part in the 9th International conference IcETRAN held in Novi Pazar, Serbia on June 6-9, 2022.
May 31st 2022
An important dissemination event within Q7 was the Workshop entitled “Trends in the development and applications of artificial intelligence”. The Workshop was held in a hybrid form, in vivo and online, at the Faculty of Electronic Engineering, University of Niš, on May 31, 2022. The Workshop presented the main trends in AI from different aspects (fundamental research, applications and commercialization). Members of the project team, experts from Serbia and abroad participated as lecturers, while there were about 40 participants in the audience (half of them in vivo and the other half online). All the lectures were recorded. The promotional materials purchased for the Workshop (cups, notebooks, pens and bags) were distributed among the presenters and participants.
March 31st 2022
On March 31, 2022 at the Faculty of Electronic Engineering, University of Niš, prof. Dr. Reinhold Häb-Umbach from the University of Paderborn in Germany gave an invited lecture entitled "Computational Analysis of Sound Scenes and Events". This lecture presented a taxonomy of tasks in the field of sound recognition and provided a discussion on generic system architecture and evaluation metrics. Attendees were also able to learn about current research challenges in detection and classification of acoustic scenes and events. Finally, further plans for cooperation of the project team members with prof. Häb-Umbach was established.
March 18th 2022
The team member Nikola Vučić took part in the International Symposium INFOTEH-JAHORINA 2022 Jahorina, East Sarajevo, Bosnia and Herzegovina, March 16-18, 2022.
The team member Jelena Nikolić took part in the online version of the International Symposium INFOTEH-JAHORINA 2022 Jahorina, East Sarajevo, Bosnia and Herzegovina, March 16-18, 2022.
The team member Nikola Vučić took part in the online version of the 7th International Mardin Artuklu Scientific Researches Conference, Mardin, Turkey, December 10-12, 2021.
February 17th-18th 2022
The most important dissemination activity during Q6 was Training school, organized as an on-line event in the period 17-18th of February 2022. The platform used for the event was MS Teams. The participation was free of charge. The immense promotion of the Training school took part over various channels: project web site - "News" tab (https://com-in-ai.elfak.rs/), SRO's student portal (https://sip.elfak.ni.ac.rs/article/konkursi/trening-skola-com-in-al-2022), mailing list of SRO’s employees, mailing list of members of Science and Technology Park Niš, Facebook page of the project (https://www.facebook.com/project.cominai) and SRO (https://www.facebook.com/elektronskifakultetnis), professional and personal contacts etc. As a result of this successful promotional campaign we have received a large number of registered participants (more than 120). We have managed to find very expertised presenters from the field of artificial intelligence for this event. Their presentations took the attention of the participants and enabled them to get new knowledge and skills from this field. All the lectures were recorded and will be available online as a precious base of knowledge. At the end of the Training school, the participants were asked to fill in the questionnaire form and to give their feedback about the event to the organizers. The promotional materials purchased for the Training school (cups, notebooks and bags) were distributed among the presenters and participants.
17. и 18. фебруар 2022.
У оквиру пројекта „Напредне методе квантизације, компресије и учења у вештачкој интелигенцији“ (Com-in-AI) који се средствима Фонда за науку Републике Србије реализује на Електронском факултету у Нишу, 17. и 18. фебруара 2022. године биће одржана оn-line тренинг школа под називом "Увод у квантизацију неуронских мрежа и примене". Главне теме тренинг школе биће:
Предавања ће у складу с Агендом држати признати стручњаци из Србије и иностранства. Учешће је бесплатно. Тренинг школа ће бити организована у виду видео конференције, док ће више детаља бити послато пријављеним учесницима.
Позивамо Вас да се пријавите путем овог линка.
December 14th 2021
Within this quarterly period, researchers from our project team took part at three conferences (TELFOR, SAUM and TELSIKS).
In September we also organized the project promotion within the SAUM conference. Namely, project leader prof. Perić and dr. Dinčić gave a presentation whereby they promoted the project Com-in-AI. They explained and summarized the project results and objectives to the visitors and pointed out the project importance for developing science in the field of AI in our country. During this event we exchanged ideas with other researchers and made new connections. To raise the reputation of the project we shared promotional materials (notebooks, pens, bags) at the project promotion at the SAUM conference.
An important meeting related to artificial intelligence and its development in Niš took place on 13th of October at Science and Technology Park of Niš whereby a recently established Artificial Intelligence Institute was presented. Many people from academia and industry attended the event, as well as students and other interested citizens. The contacts with the AI Institute and its leaders are strengthened. Project leader prof. Perić made a short oral presentation of the Com-in-AI project to the audience. The promotional materials were distributed to the visitors and the contacts with potential partners for future collaboration were exchanged.
September 15th 2021
The team member Nikola Vučić took part in the online version of the 25th International conference ELECTRONICS 2021, where he presented a scientific paper selected for publication in the scientific journal Elektronika Ir Elektrotechnika:
Zoran Perić, Bojan Denić, Milan Savić, Nikola Vučić and Nikola Simić, "Binary quantization analysis of neural networks weights on MNIST dataset", Elektronika Ir Elektrotechnika, vol. 27, no. 4, pp. 55-61, 2021. https://doi.org/10.5755/j02.eie.28881
June 13th 2021
Team members P1 (prof. Ćirić) and P2 (prof. Jovanović) took part in the online version (due to the covid-19 pandemic) of 20th International Symposium INFOTEH-JAHORINA (INFOTEH).
Team members P5 (Nikola Vučić) took part in the online version of 12th International Symposium on Advanced Topics in Electrical Engineering (ATEE), where he presented a scientific paper.
Team member P1 (prof. Ćirić) participated in the online event “AI meetup: Artificial intelligence – strategy, conditions and challenges” held on the 20th of April 2021, where he gave a presentation promoting the project Com-in-AI.
October 9th 2020
The project Com-in-AI was presented on October 9th 2020 at Kalemegdan (Belgrade) within the Science Fund exposition dedicated to the promotion of the Program for the development of projects in the domain of AI. More details at link.
The dissemination, exploitation and communication activities will be carried out continuously throughout the lifetime of the project, but also after the project completion, to ensure long-term effects of the project. The dissemination activities will be focused on promoting the scientific results, the potential usage and also the project in general among a wide scientific community as well as among the target groups relevant for exploitation of the achieved results. Workshops to be organized will target the scientific and the interested non-scientific stakeholders (industry, AI companies, Public authorities at national and local levels, relevant decision makers) to exchange ideas on the scientific, economic and social aspects as well as applicability of the project results and to discuss possible collaboration.
Here you can find a list of sorted literature relevant to the project topic:
>>> Link to the list of relevant literature <<<
It is well known that deep learning becomes the most profitable when applied to large training datasets. This is why audio data from industrial machines/products have been acquired and recorded during the course of the project. For the purpose of machine condition monitoring, among others, sounds of direct current (DC) motors and pumps from the home heating/air conditioning systems were recorded and stored. The acquired sounds have been used for implementing theoretical solutions on the developed DNNs, as a part of the target use case (prediction of the working condition of a tested machine).
Sounds of DC motors: Two different types of DC motors (type A and B) were recorded. All files were stored as mono audio in wav format, with sampling frequency of 16 kHz, and of duration of 18 s. There are 6591 files for the motor A, and 7343 files for the motor B. The overall size of the dataset is about 7.5 GB.
https://drive.google.com/drive/u/2/folders/1bGtR45gY5KoNmg4NR_3nD_KNNX36m3Ae
Sounds of water pumps: Sounds of two water pumps (good and bad) were measured, where the measurements were done using the condenser measurement microphone of class 1. The dataset consists of 2479 (for good pump) and 2483 (for bad pump) mono audio files in wav format, and its size is about 2 GB. Duration of each audio signal is 5 s, while the sampling frequency is 44.1 kHz.
https://drive.google.com/drive/u/2/folders/1NcUtjQX-M4U7Fm9FrP46nZrJsxQrnXXk
This research was supported by the Science Fund of the Republic of Serbia, 6527104, AI- Com-in-AI.
This website is made with the financial support of the Science Fund of the Republic of Serbia.
For the website's content is only responsible University of Niš, Faculty of Electronic Engineering and the content doesn't represent the attitudes of the Science Fund of the Republic of Serbia.