Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158637
Cristian Avramescu, A. Tenescu, B. Bercean, Marius Marcu
Classification and object detection are computer vision tasks with successful, clinical applications in medical imaging. Yet, the increased effort required of expert readers in order to annotate bounding boxes on medical images has yet to be quantitatively justified in terms of added value to identifying pathologies. In this study, we show preliminary results on the classification and localization of 17 most common chest pathologies on a private dataset of 15,000 radiographs from two Romanian public hospitals. Next, we quantitatively compare the extra added value of the bounding box information at training time, concerning classification performance improvements. Two types of architectures were trained on publicly available and private data, classification architectures (i.e., InceptionNet V3), tasked with identifying pathologies in chest radiographies and object detection architectures (i.e., Faster R-CNN), tasked with localizing the regions of interest on the image. Both achieved high classification performance (i.e., 90.52 and 88.94 mean AUROC, respectively). The object detector, however, reached superior classification performance, thus proving the additional bounding box information available at training time, benefits patient-level pathology identification as well.
{"title":"Bounding Box Supervision Benefits Lung Pathology Classification in Pulmonary X-Rays","authors":"Cristian Avramescu, A. Tenescu, B. Bercean, Marius Marcu","doi":"10.1109/SACI58269.2023.10158637","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158637","url":null,"abstract":"Classification and object detection are computer vision tasks with successful, clinical applications in medical imaging. Yet, the increased effort required of expert readers in order to annotate bounding boxes on medical images has yet to be quantitatively justified in terms of added value to identifying pathologies. In this study, we show preliminary results on the classification and localization of 17 most common chest pathologies on a private dataset of 15,000 radiographs from two Romanian public hospitals. Next, we quantitatively compare the extra added value of the bounding box information at training time, concerning classification performance improvements. Two types of architectures were trained on publicly available and private data, classification architectures (i.e., InceptionNet V3), tasked with identifying pathologies in chest radiographies and object detection architectures (i.e., Faster R-CNN), tasked with localizing the regions of interest on the image. Both achieved high classification performance (i.e., 90.52 and 88.94 mean AUROC, respectively). The object detector, however, reached superior classification performance, thus proving the additional bounding box information available at training time, benefits patient-level pathology identification as well.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128365787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158575
Ámon Kiss, András Kovács
This paper presents a solution to the problem of real-time food detection on mobile devices. The solution compares a YOLOv4 tiny model and a multilayer classification neural network ensemble as the convolutional neural network responsible for food detection, using MobilenetV3 for classification and several YOLOv4 tiny models capable of detecting less food in the overall problem. The model was trained with 40 different food categories using an RTX 3070 video card. Results showed that the multilayer network ensemble performed better than a single YOLOv4 tiny model. The TP value was 258 higher, the FP value was 163 lower and the FN value was 773 lower. The accuracy increased from 0,56 to 0,61, the recall increased from 0,52 to 0,70 and the F1-score also increased from 0,54 to 0,65. Finally, the finished models were implemented on an Android mobile device, and the application allows users to log meals and generate statistics from the saved data.
{"title":"Food recognition using neural network on mobile device","authors":"Ámon Kiss, András Kovács","doi":"10.1109/SACI58269.2023.10158575","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158575","url":null,"abstract":"This paper presents a solution to the problem of real-time food detection on mobile devices. The solution compares a YOLOv4 tiny model and a multilayer classification neural network ensemble as the convolutional neural network responsible for food detection, using MobilenetV3 for classification and several YOLOv4 tiny models capable of detecting less food in the overall problem. The model was trained with 40 different food categories using an RTX 3070 video card. Results showed that the multilayer network ensemble performed better than a single YOLOv4 tiny model. The TP value was 258 higher, the FP value was 163 lower and the FN value was 773 lower. The accuracy increased from 0,56 to 0,61, the recall increased from 0,52 to 0,70 and the F1-score also increased from 0,54 to 0,65. Finally, the finished models were implemented on an Android mobile device, and the application allows users to log meals and generate statistics from the saved data.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121175274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158663
Eszter Virágh, D. Drexler, B. Kiss
For several chemical reactions, the intermediate states of the system has to be known. They can be given, e.g., by sampling, measurements, or motion planning. By motion planning, we can determine the time function of the concentrations and the temperature, using the mathematical model of the system. From this, we can conclude the state of the system at every moment. We provide a motion planning method for isothermal parallel structural chemical reactions. We simulate the behavior of the three-step parallel reaction. The results of the motion planning can help to decide, if it is necessary to intervene from outside, thus we can ensure that the concentration is in a specified interval.
{"title":"Motion planning and modeling for isothermal parallel chemical reactions","authors":"Eszter Virágh, D. Drexler, B. Kiss","doi":"10.1109/SACI58269.2023.10158663","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158663","url":null,"abstract":"For several chemical reactions, the intermediate states of the system has to be known. They can be given, e.g., by sampling, measurements, or motion planning. By motion planning, we can determine the time function of the concentrations and the temperature, using the mathematical model of the system. From this, we can conclude the state of the system at every moment. We provide a motion planning method for isothermal parallel structural chemical reactions. We simulate the behavior of the three-step parallel reaction. The results of the motion planning can help to decide, if it is necessary to intervene from outside, thus we can ensure that the concentration is in a specified interval.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"64 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125948157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158544
Patrik Viktor, Mónika Garai-Fodor
This paper deals with a generation-specific approach to adaptive devices for self-driving systems in Hungary. In the literature section, we discuss self-driving and generation-specific. In the primary research, we conduct a factor analysis based on a questionnaire survey. We received 590 responses to the questionnaire, which were analyzed using SPSS software. The study analysed attitudes towards self-driving cars from a generation-specific perspective. The results of the study clearly demonstrated that the generations, differentiated in their values and mindsets, have different attitudes towards self-driving cars. We believe that the results can help to identify the specificities of the target group for self-driving vehicles and to refine potential target groups.
{"title":"Generation-Specific Analysis of Adaptive Self-Driving Technology in Hungary","authors":"Patrik Viktor, Mónika Garai-Fodor","doi":"10.1109/SACI58269.2023.10158544","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158544","url":null,"abstract":"This paper deals with a generation-specific approach to adaptive devices for self-driving systems in Hungary. In the literature section, we discuss self-driving and generation-specific. In the primary research, we conduct a factor analysis based on a questionnaire survey. We received 590 responses to the questionnaire, which were analyzed using SPSS software. The study analysed attitudes towards self-driving cars from a generation-specific perspective. The results of the study clearly demonstrated that the generations, differentiated in their values and mindsets, have different attitudes towards self-driving cars. We believe that the results can help to identify the specificities of the target group for self-driving vehicles and to refine potential target groups.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132219196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158668
Eniko Nagy, Ildikó Holik
Today’s technological developments are also having an impact on education. In higher education, there are more and more innovations in content and methodology, such as educational robots, which can be used not only for teaching programming, but also as teaching assistants and pedagogical assistants. Tutor robots offer new opportunities for curriculum development, motivating students, stimulating interest and developing soft skills, while preparing young people to use the latest technological tools. In our online survey, conducted in summer 2022, we explored the views of educators in 14 countries on the characteristics of robot-supported innovation in higher education. The results of the survey show that educators are open to educational innovations. Educational robots are mainly used for teaching programming. They believe that the use of social robots significantly improves students’ creativity and self-expression. The results of this research have shown that the use of educational social robots is an excellent opportunity for methodological innovation and provides scope for experimental teaching.
{"title":"Educational robots in higher education -findings from an international survey","authors":"Eniko Nagy, Ildikó Holik","doi":"10.1109/SACI58269.2023.10158668","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158668","url":null,"abstract":"Today’s technological developments are also having an impact on education. In higher education, there are more and more innovations in content and methodology, such as educational robots, which can be used not only for teaching programming, but also as teaching assistants and pedagogical assistants. Tutor robots offer new opportunities for curriculum development, motivating students, stimulating interest and developing soft skills, while preparing young people to use the latest technological tools. In our online survey, conducted in summer 2022, we explored the views of educators in 14 countries on the characteristics of robot-supported innovation in higher education. The results of the survey show that educators are open to educational innovations. Educational robots are mainly used for teaching programming. They believe that the use of social robots significantly improves students’ creativity and self-expression. The results of this research have shown that the use of educational social robots is an excellent opportunity for methodological innovation and provides scope for experimental teaching.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133988264","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158552
Ágnes Csiszárik-Kocsir, J. Varga
The events of recent years have shed new light on the crisis and change management practices of economic actors. It has been a truism that actors who are more receptive to change can be more successful and efficient than their peers in business markets. Perhaps the most significant changes in recent years have confirmed this even more. One only has to think of the impact of pandemic COVID-19, the energy crisis or the Russian-Ukrainian conflict. In a very short period of time, these events have brought about very significant changes and their impact has been largely negative for most economic actors. If it has not been sufficiently understood so far why it can be important to adapt to changes in a timely and appropriate way, or why good crisis management practices can be important, perhaps everyone will now. Meanwhile, other trends are shaping the global economy and will have an impact on the future state of the economy and society. Examples include sustainability (or the green transition) and digitalisation. Two global changes that will certainly have a long-term impact on society and business processes. No one can afford the luxury of ignoring these changes. What is more, the most competitive economic players are seeking to turn them to their advantage and to reap the benefits of sustainability or digitalisation. It is clear that our world has become faster and more complex than ever before. More and more things are changing around us, with ever more intense consequences. We need to recognise in time how we can respond to changing environmental conditions or circumstances. This is the subject of the present paper, which, after a brief literature review, draws on research findings to illustrate the importance and relevance of digitalisation.
{"title":"The advancing role of digitalisation through the example of the Perlmutter project from the user side","authors":"Ágnes Csiszárik-Kocsir, J. Varga","doi":"10.1109/SACI58269.2023.10158552","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158552","url":null,"abstract":"The events of recent years have shed new light on the crisis and change management practices of economic actors. It has been a truism that actors who are more receptive to change can be more successful and efficient than their peers in business markets. Perhaps the most significant changes in recent years have confirmed this even more. One only has to think of the impact of pandemic COVID-19, the energy crisis or the Russian-Ukrainian conflict. In a very short period of time, these events have brought about very significant changes and their impact has been largely negative for most economic actors. If it has not been sufficiently understood so far why it can be important to adapt to changes in a timely and appropriate way, or why good crisis management practices can be important, perhaps everyone will now. Meanwhile, other trends are shaping the global economy and will have an impact on the future state of the economy and society. Examples include sustainability (or the green transition) and digitalisation. Two global changes that will certainly have a long-term impact on society and business processes. No one can afford the luxury of ignoring these changes. What is more, the most competitive economic players are seeking to turn them to their advantage and to reap the benefits of sustainability or digitalisation. It is clear that our world has become faster and more complex than ever before. More and more things are changing around us, with ever more intense consequences. We need to recognise in time how we can respond to changing environmental conditions or circumstances. This is the subject of the present paper, which, after a brief literature review, draws on research findings to illustrate the importance and relevance of digitalisation.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125217488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158647
M. Kucarov, A. Várkonyi-Kóczy, B. Molnár, M. Kozlovszky
Such laser microdissection design is discussed that requires no human interaction and yet capable of μm accuracy. The goal is to boost speed, accuracy, and smooth workflow between slide scanning and laser cutting of suspicious areas from tissue sample based on software evaluation of the digital slides. Fully automated, precise, robust solution has been developed using hybrid method that includes static matrix transformation for preliminary position calculation, computer vision, and image processing technique for real-time object detection and position adjustment. Software implementation also tested on microdissector equipment. Applying this methodology in clinical pathology could reduce the waiting lists, the overload of clinics, laboratory technicians.
{"title":"Single Cell Position Determination and Transformation from Static High-resolution Digital Image to Laser-microdissector Coordinate System Using Image Processing Techniques","authors":"M. Kucarov, A. Várkonyi-Kóczy, B. Molnár, M. Kozlovszky","doi":"10.1109/SACI58269.2023.10158647","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158647","url":null,"abstract":"Such laser microdissection design is discussed that requires no human interaction and yet capable of μm accuracy. The goal is to boost speed, accuracy, and smooth workflow between slide scanning and laser cutting of suspicious areas from tissue sample based on software evaluation of the digital slides. Fully automated, precise, robust solution has been developed using hybrid method that includes static matrix transformation for preliminary position calculation, computer vision, and image processing technique for real-time object detection and position adjustment. Software implementation also tested on microdissector equipment. Applying this methodology in clinical pathology could reduce the waiting lists, the overload of clinics, laboratory technicians.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131019757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158621
Lehel Dénes-Fazakas, M. Siket, László Szilágyi, G. Eigner, L. Kovács
In the present study, we investigated the effect of different reward functions in insulin regulation using reinforcement learning. An artificial pancreas system is able to deliver insulin into the body in an automated way. The control algorithm of an automated insulin delivery system is a key player in achieving personalized therapy. Neural networks provide an approach to customize insulin administration by learning the patient’s habits and administering insulin accordingly. Therefore, we conducted experiments with neural networks based on reinforcement learning. Our goal was to find a neural network-based model and reward function that could learn the patient’s behavior and administers insulin with the best time in ranges. We evaluated the method on simulated virtual patients when sensor noise occurs. The results show that the bump functions were the most efficient in providing acceptable time in ranges.
{"title":"Investigation of reward functions for controlling blood glucose level using reinforcement learning","authors":"Lehel Dénes-Fazakas, M. Siket, László Szilágyi, G. Eigner, L. Kovács","doi":"10.1109/SACI58269.2023.10158621","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158621","url":null,"abstract":"In the present study, we investigated the effect of different reward functions in insulin regulation using reinforcement learning. An artificial pancreas system is able to deliver insulin into the body in an automated way. The control algorithm of an automated insulin delivery system is a key player in achieving personalized therapy. Neural networks provide an approach to customize insulin administration by learning the patient’s habits and administering insulin accordingly. Therefore, we conducted experiments with neural networks based on reinforcement learning. Our goal was to find a neural network-based model and reward function that could learn the patient’s behavior and administers insulin with the best time in ranges. We evaluated the method on simulated virtual patients when sensor noise occurs. The results show that the bump functions were the most efficient in providing acceptable time in ranges.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114859582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158538
Florin Butaru, Mihaela-Codruta Ancuti, G. Erdodi, C. Șorândaru, S. Musuroi, R. Ancuti
In this paper, the operation of the wind turbine (WT) at its maximum power point (MPP) is analyzed, using the classical perturb and observe method (POM). The wind system, consisting of the wind turbine (WT) and the electric generator, is currently operating in the energetically optimal zone using the perturb and observe method (POM) at constant wind speeds over time. This research demonstrates the inefficiency of the perturb and observe method (POM) at time-varying wind speeds using extended simulation findings based on experimental data from the wind turbines region in Dobrogea and using mathematical models.
{"title":"Wind System Control at Time-Varying Wind Speeds Using the Perturb and Observe Method","authors":"Florin Butaru, Mihaela-Codruta Ancuti, G. Erdodi, C. Șorândaru, S. Musuroi, R. Ancuti","doi":"10.1109/SACI58269.2023.10158538","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158538","url":null,"abstract":"In this paper, the operation of the wind turbine (WT) at its maximum power point (MPP) is analyzed, using the classical perturb and observe method (POM). The wind system, consisting of the wind turbine (WT) and the electric generator, is currently operating in the energetically optimal zone using the perturb and observe method (POM) at constant wind speeds over time. This research demonstrates the inefficiency of the perturb and observe method (POM) at time-varying wind speeds using extended simulation findings based on experimental data from the wind turbines region in Dobrogea and using mathematical models.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117124262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-23DOI: 10.1109/SACI58269.2023.10158603
Henrik Varga, A. Kiss, Zoltán Kolarovszki
In the past few years, quantum computing has gotten more attention, and the need for efficient simulations is getting increasingly important as well. A significant branch of quantum computing is photonic quantum computing. For simulating photonic quantum circuits, Strawberry Fields is the most popular framework. In this paper, we compared it with another framework currently under development called Piquasso regarding gradient calculation time, which is an essential part of continuous-variable quantum neural networks. We present the apparent scalability of Piquasso over Strawberry Fields by storing fewer data, but leading to possible accuracy differences as a trade-off, which could motivate future work.
{"title":"Benchmarking Photonic Quantum Machine Learning Simulators","authors":"Henrik Varga, A. Kiss, Zoltán Kolarovszki","doi":"10.1109/SACI58269.2023.10158603","DOIUrl":"https://doi.org/10.1109/SACI58269.2023.10158603","url":null,"abstract":"In the past few years, quantum computing has gotten more attention, and the need for efficient simulations is getting increasingly important as well. A significant branch of quantum computing is photonic quantum computing. For simulating photonic quantum circuits, Strawberry Fields is the most popular framework. In this paper, we compared it with another framework currently under development called Piquasso regarding gradient calculation time, which is an essential part of continuous-variable quantum neural networks. We present the apparent scalability of Piquasso over Strawberry Fields by storing fewer data, but leading to possible accuracy differences as a trade-off, which could motivate future work.","PeriodicalId":339156,"journal":{"name":"2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121934390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：