Journal Description
Computer Sciences & Mathematics Forum
Computer Sciences & Mathematics Forum
is an open access journal dedicated to publishing findings resulting from academic conferences, workshops, and similar events in the area of computer science and mathematics. Each conference proceeding can be individually indexed, is citable via a digital object identifier (DOI), and is freely available under an open access license. The conference organizers and proceedings editors are responsible for managing the peer-review process and selecting papers for conference proceedings.
Latest Articles
Overview of Training LLMs on One Single GPU
Comput. Sci. Math. Forum 2025, 10(1), 14; https://doi.org/10.3390/cmsf2025010014 - 9 Jul 2025
Abstract
Large language models (LLMs) are developing at a rapid pace, which has made it necessary to better understand how they train, especially when faced with resource limitations. This paper examines in detail how various state-of-the-art LLMs train on a single Graphical Processing Unit
[...] Read more.
Large language models (LLMs) are developing at a rapid pace, which has made it necessary to better understand how they train, especially when faced with resource limitations. This paper examines in detail how various state-of-the-art LLMs train on a single Graphical Processing Unit (GPU), paying close attention to crucial elements like throughput, memory utilization and training time. We find important trade-offs between model size, batch size and computational efficiency through empirical evaluation, offering practical advice for streamlining fine-tuning processes in the face of hardware constraints.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►
Show Figures
Open AccessProceeding Paper
Optimizing Machine Learning for Healthcare Applications: A Case Study on Cardiovascular Disease Prediction Through Feature Selection, Regularization, and Overfitting Reduction
by
Lamiae Eloutouate, Hicham Gibet Tani, Lotfi Elaachak, Fatiha Elouaai and Mohammed Bouhorma
Comput. Sci. Math. Forum 2025, 10(1), 13; https://doi.org/10.3390/cmsf2025010013 - 7 Jul 2025
Abstract
The application of machine learning (ML) to medical datasets offers significant potential for improving disease prediction and patient outcomes. However, challenges such as feature redundancy, overfitting, and suboptimal model performance limit the practical effectiveness of ML algorithms. This study focuses on optimizing ML
[...] Read more.
The application of machine learning (ML) to medical datasets offers significant potential for improving disease prediction and patient outcomes. However, challenges such as feature redundancy, overfitting, and suboptimal model performance limit the practical effectiveness of ML algorithms. This study focuses on optimizing ML techniques for cardiovascular disease prediction using the Kaggle Cardiovascular Disease dataset. We systematically apply feature selection methods, including correlation analysis and regularization techniques (L1/L2), to identify the most relevant attributes and address multicollinearity. Advanced ensemble models such as Random Forest, XGBoost, and LightGBM are employed to mitigate overfitting and enhance predictive performance. Through hyperparameter tuning and stratified k-fold cross-validation, we ensure model robustness and generalizability. The results demonstrate that ensemble methods, particularly gradient boosting algorithms, outperform traditional models, achieving superior predictive accuracy and stability. This study highlights the importance of algorithm optimization in ML applications for healthcare, offering a replicable framework for medical datasets and paving the way for more effective diagnostic tools in cardiovascular health.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Integrating Machine Learning with Medical Imaging for Human Disease Diagnosis: A Survey
by
Anass Roman, Chaymae Taib, Ilham Dhaiouir and Haimoudi El Khatir
Comput. Sci. Math. Forum 2025, 10(1), 12; https://doi.org/10.3390/cmsf2025010012 - 7 Jul 2025
Abstract
Machine learning is revolutionizing healthcare by enhancing diagnosis and treatment personalization. This study explores ML applications in medical imaging, analyzing data from X-rays, CT, MRI, and ultrasound for early disease detection. It reviews key ML models, including SVM, ANN, RF, CNN, and other
[...] Read more.
Machine learning is revolutionizing healthcare by enhancing diagnosis and treatment personalization. This study explores ML applications in medical imaging, analyzing data from X-rays, CT, MRI, and ultrasound for early disease detection. It reviews key ML models, including SVM, ANN, RF, CNN, and other methods, demonstrating their effectiveness in detecting cancers such as lung and prostate cancer and other diseases. Despite their accuracy, these methods face challenges such as a reliance on large datasets and significant computational requirements. This study highlights the need for further research to integrate ML into clinical practice, addressing its limitations and unlocking new opportunities for improved patient care.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
The Evolution and Challenges of Real-Time Big Data: A Review
by
Ikram Lefhal Lalaoui, Essaid El Haji and Mohamed Kounaidi
Comput. Sci. Math. Forum 2025, 10(1), 11; https://doi.org/10.3390/cmsf2025010011 - 1 Jul 2025
Abstract
The importance of real-time big data has become crucial in the digital revolution of modern society, in the context of increasing data flows from multiple sources, including social media, internet connected devices (IOT) and financial systems, real-time analysis and processing is becoming a
[...] Read more.
The importance of real-time big data has become crucial in the digital revolution of modern society, in the context of increasing data flows from multiple sources, including social media, internet connected devices (IOT) and financial systems, real-time analysis and processing is becoming a strategic tool for fast and accurate decision making, we find applications in different domains such as healthcare, finance, and digital marketing, which is revolutionizing traditional business models. In this article, we explore the recent advances and future prospects of real-time big data. Our research is based on recent work published between 2020 and 2025, examining the technological advances, the difficulties encountered and suggesting ways of optimizing the efficiency of these technologies.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
Open AccessProceeding Paper
Advancing Stress Detection and Health Monitoring with Deep Learning Approaches
by
Merouane Mouadili, El Mokhtar En-Naimi and Mohamed Kouissi
Comput. Sci. Math. Forum 2025, 10(1), 10; https://doi.org/10.3390/cmsf2025010010 - 1 Jul 2025
Abstract
Numerous studies in the healthcare field conducted in recent years have highlighted the impact of stress on health and its role in the development of several critical illnesses. Stress monitoring using wearable technologies, such as smartwatches and biosensors, has shown promising results in
[...] Read more.
Numerous studies in the healthcare field conducted in recent years have highlighted the impact of stress on health and its role in the development of several critical illnesses. Stress monitoring using wearable technologies, such as smartwatches and biosensors, has shown promising results in improving the management of this issue. Data from both physical and mental health can be leveraged to enhance medical decision-making, support research on new treatments, and deepen our understanding of complex diseases. However, traditional machine learning (ML) systems often face limitations, particularly in real-time processing and resource optimization, which restrict their application in critical situations. In this article, we present the development of a deep learning-based approach that leverages models such as 1D CNN (Convolutional Neural Networks), LSTM (Long Short-Term Memory), and Time-Series Transformers, alongside classical deep learning techniques. We then highlight the transformative potential of TinyML for real-time, low-power health monitoring, focusing on Heart Rate Variability (HRV) analysis. This approach aims to optimize personalized health interventions and enhance the accuracy of medical monitoring.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Generative AI Respiratory and Cardiac Sound Separation Using Variational Autoencoders (VAEs)
by
Arshad Jamal, R. Kanesaraj Ramasamy and Junaidi Abdullah
Comput. Sci. Math. Forum 2025, 10(1), 9; https://doi.org/10.3390/cmsf2025010009 - 1 Jul 2025
Abstract
The separation of respiratory and cardiac sounds is a significant challenge in biomedical signal processing due to their overlapping frequency and time characteristics. Traditional methods struggle with accurate extraction in noisy or diverse clinical environments. This study explores the application of machine learning,
[...] Read more.
The separation of respiratory and cardiac sounds is a significant challenge in biomedical signal processing due to their overlapping frequency and time characteristics. Traditional methods struggle with accurate extraction in noisy or diverse clinical environments. This study explores the application of machine learning, particularly convolutional neural networks (CNNs), to overcome these obstacles. Advanced machine learning models, denoising algorithms, and domain adaptation strategies address challenges such as frequency overlap, external noise, and limited labeled datasets. This study presents a robust methodology for detecting heart and lung diseases from audio signals using advanced preprocessing, feature extraction, and deep learning models. The approach integrates adaptive filtering and bandpass filtering as denoising techniques and variational autoencoders (VAEs) for feature extraction. The extracted features are input into a CNN, which classifies audio signals into different heart and lung conditions. The results highlight the potential of this combined approach for early and accurate disease detection, contributing to the development of reliable diagnostic tools for healthcare.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Enhanced Lung Disease Detection Using Double Denoising and 1D Convolutional Neural Networks on Respiratory Sound Analysis
by
Reshma Sreejith, R. Kanesaraj Ramasamy, Wan-Noorshahida Mohd-Isa and Junaidi Abdullah
Comput. Sci. Math. Forum 2025, 10(1), 7; https://doi.org/10.3390/cmsf2025010007 - 24 Jun 2025
Abstract
The accurate and early detection of respiratory diseases is vital for effective diagnosis and treatment. This study presents a new approach for classifying lung sounds using a double denoising method combined with a 1D Convolutional Neural Network (CNN). The preprocessing uses Fast Fourier
[...] Read more.
The accurate and early detection of respiratory diseases is vital for effective diagnosis and treatment. This study presents a new approach for classifying lung sounds using a double denoising method combined with a 1D Convolutional Neural Network (CNN). The preprocessing uses Fast Fourier Transform to clean up sounds and High-Pass Filtering to improve the quality of breathing sounds by eliminating noise and low-frequency interruptions. The Short-Time Fourier Transform (STFT) extracts features that capture localised frequency variations, crucial for distinguishing normal and abnormal respiratory sounds. These features are input into the 1D CNN, which classifies diseases such as bronchiectasis, pneumonia, asthma, COPD, healthy, and URTI. The dual denoising method enhances signal clarity and classification performance. The model achieved 96% validation accuracy, highlighting its reliability in detecting respiratory conditions. The results emphasise the effectiveness of combining signal augmentation with deep learning for automated respiratory sound analysis, with future research focusing on dataset expansion and model refinement for clinical use.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Blockchain for Sustainable Smart Cities: Motivations and Challenges
by
Fatima Zahrae Chentouf, Mohamed El Alami Hasoun and Said Bouchkaren
Comput. Sci. Math. Forum 2025, 10(1), 2; https://doi.org/10.3390/cmsf2025010002 - 17 Jun 2025
Abstract
Rapid urbanization and the rising demand for sustainable living have encouraged the growth of smart cities, which incorporate innovative technologies to ameliorate environmental sustainability, optimize resource management, and improve living standards. The convergence of blockchain (BC) technology and the Internet of Things (IoT)
[...] Read more.
Rapid urbanization and the rising demand for sustainable living have encouraged the growth of smart cities, which incorporate innovative technologies to ameliorate environmental sustainability, optimize resource management, and improve living standards. The convergence of blockchain (BC) technology and the Internet of Things (IoT) presents transformative convenience for managing smart cities and achieving sustainability goals. In fact, blockchain technology combined with IoT devices provides a decentralized, transparent, and safe framework for managing massive volumes of data produced by networked sensors and systems. By guaranteeing accountability, minimizing fraud, and maximizing resource use, blockchain not only facilitates the smooth operation of smart city infrastructures but also encourages sustainable habits. The various uses of blockchain technology in smart city management and its contribution to sustainability objectives are examined in this study. Through an examination of important domains like energy distribution, waste management, transportation systems, healthcare, and governance, the research shows how blockchain promotes effective data exchange and data security, builds stakeholder trust, and makes it possible to establish decentralized organizations to improve decision-making.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessEditorial
Preface of Sustainable Computing and Green Technologies (SCGT’2025)
by
Hicham Gibet Tani, Mohamed Kouissi, Mohamed Ben Ahmed, Anouar Boudhir Abdelhakim and Lotfi Elaachak
Comput. Sci. Math. Forum 2025, 10(1), 1; https://doi.org/10.3390/cmsf2025010001 - 17 Jun 2025
Abstract
n/a
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
Open AccessProceeding Paper
Enhancing Security and Privacy in IoT Data Streams: Real-Time Anomaly Detection for Threat Mitigation in Traffic Management
by
Oumayma Berraadi, Hicham Gibet Tani and Mohamed Ben Ahmed
Comput. Sci. Math. Forum 2025, 10(1), 8; https://doi.org/10.3390/cmsf2025010008 - 16 Jun 2025
Abstract
The rapid expansion of IoT in smart cities has improved traffic management but increased security risks. Traditional IDS struggle with advanced threats, prompting adaptive solutions. This work proposes a framework combining machine learning (ML), Zero Trust Architecture (ZTA), and blockchain authentication. Supervised models
[...] Read more.
The rapid expansion of IoT in smart cities has improved traffic management but increased security risks. Traditional IDS struggle with advanced threats, prompting adaptive solutions. This work proposes a framework combining machine learning (ML), Zero Trust Architecture (ZTA), and blockchain authentication. Supervised models (XGBoost, RF, SVM, LR) detect known anomalies, while a CNN Autoencoder identifies novel threats. Blockchain ensures identity integrity, and compromised devices are isolated automatically. Tests on the IoT-23 dataset demonstrate superior accuracy, fewer false positives, and better scalability than conventional methods. The integration of AI, Zero Trust, and blockchain significantly boosts IoT traffic system security and resilience.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Comparative Analysis of Energy Consumption and Carbon Footprint in Automatic Speech Recognition Systems: A Case Study Comparing Whisper and Google Speech-to-Text
by
Jalal El Bahri, Mohamed Kouissi and Mohammed Achkari Begdouri
Comput. Sci. Math. Forum 2025, 10(1), 6; https://doi.org/10.3390/cmsf2025010006 - 16 Jun 2025
Abstract
This study investigates the energy consumption and carbon footprint of two prominent automatic speech recognition (ASR) systems: OpenAI’s Whisper and Google’s Speech-to-Text API. We evaluate both local and cloud-based speech recognition approaches using a public Kaggle dataset of 20,000 short audio clips in
[...] Read more.
This study investigates the energy consumption and carbon footprint of two prominent automatic speech recognition (ASR) systems: OpenAI’s Whisper and Google’s Speech-to-Text API. We evaluate both local and cloud-based speech recognition approaches using a public Kaggle dataset of 20,000 short audio clips in Urdu, utilizing CodeCarbon, PyJoule, and PowerAPI for comprehensive energy profiling. As a result of our analysis, we expose some substantial differences between the two systems in terms of energy efficiency and carbon emissions, with the cloud-based solution showing substantially lower environmental impact despite comparable accuracy. We discuss the implications of these findings for sustainable AI deployment and minimizing the ecological footprint of speech recognition technologies.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
Open AccessProceeding Paper
Green DevOps: A Strategic Framework for Sustainable Software Development
by
Mohamed Toufik Ailane, Carolin Rubner and Andreas Rausch
Comput. Sci. Math. Forum 2025, 10(1), 5; https://doi.org/10.3390/cmsf2025010005 - 16 Jun 2025
Abstract
In the era of digital transformation, the environmental impact of software development has become a pressing concern, necessitating the integration of sustainability into software development processes. This paper addresses how DevOps, traditionally celebrated for enhancing efficiency and speed in software delivery, can integrate
[...] Read more.
In the era of digital transformation, the environmental impact of software development has become a pressing concern, necessitating the integration of sustainability into software development processes. This paper addresses how DevOps, traditionally celebrated for enhancing efficiency and speed in software delivery, can integrate sustainability principles to mitigate environmental impacts. We propose guidelines for integrating sustainability throughout the DevOps life cycle, aiming for significant carbon footprint reduction without compromising quality. Using a Life Cycle Assessment (LCA) approach, this study enables stakeholders to incorporate green guidelines at various software development and operation stages, enhancing software environmental friendliness. Our model supports sustainable software development and encourages proactive environmental impact minimization.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Fuzzy Logic Detector for Photovoltaic Fault Diagnosis
by
Chaymae Abdellaoui and Youssef Lagmich
Comput. Sci. Math. Forum 2025, 10(1), 4; https://doi.org/10.3390/cmsf2025010004 - 16 Jun 2025
Abstract
The performance degradation of photovoltaic (PV) systems, comprising solar panels and DC-DC converters, is often caused by various anomalies related to manufacturing defects, operational conditions, or environmental factors. These faults significantly reduce energy output, preventing the system from reaching its nominal power and
[...] Read more.
The performance degradation of photovoltaic (PV) systems, comprising solar panels and DC-DC converters, is often caused by various anomalies related to manufacturing defects, operational conditions, or environmental factors. These faults significantly reduce energy output, preventing the system from reaching its nominal power and expected production levels. Given the demonstrated impact of such faults on PV system efficiency, an effective diagnostic method is essential for proactive maintenance and optimal performance. This paper presents a fault detection algorithm based on a Mamdani-type fuzzy logic approach. The proposed method utilizes three key inputs—panel current, panel voltage, and converter voltage—to assess system health. By computing the distortion ratios of these electrical parameters and processing them through a fuzzy logic controller, the algorithm accurately identifies fault conditions. Simulation results validate the effectiveness of this approach, demonstrating its capability to detect and classify 12 distinct faults in both the PV array and the DC-DC converter. The study highlights the potential of fuzzy logic-based diagnostics in enhancing the reliability and maintenance of photovoltaic systems.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Systematic Review of AI-Driven Personalization in Serious Games for Teaching at the Right Level in Morocco
by
Najlae Abarghache, Younès Alaoui Soulimani, Lotfi Elaachak and Abderrahim Ghadi
Comput. Sci. Math. Forum 2025, 10(1), 3; https://doi.org/10.3390/cmsf2025010003 - 16 Jun 2025
Abstract
Digitalization has shaped every part of our world, from education to healthcare, from finance to entertainment, and from manufacturing to social interactions. This digital transformation has put tremendous power in the hands of individuals and delivered even better results than sticking to old,
[...] Read more.
Digitalization has shaped every part of our world, from education to healthcare, from finance to entertainment, and from manufacturing to social interactions. This digital transformation has put tremendous power in the hands of individuals and delivered even better results than sticking to old, traditional ways. Such effective digitalization cannot be achieved easily; a good digitized system is the product of a lot of effort. In this paper, we explore the state-of-the-art advancements in smart education technologies, analyze existing digital solutions, and outline considerations for developing a tailored system to digitalize the Teaching at the Right Level (TaRL) assessment for Moroccan primary education. Our approach aims to bridge skill gaps by integrating serious games, adaptive learning techniques, and real-time analytics to enhance assessment effectiveness and ease educators’ workloads.
Full article
(This article belongs to the Proceedings of International Conference on Sustainable Computing and Green Technologies (SCGT’2025))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Using Reconfigurable Multi-Valued Logic Operators to Build a New Encryption Technology
by
Hongjian Wang, Shan Ouyang, Xunlei Chen and Yi Jin
Comput. Sci. Math. Forum 2023, 8(1), 99; https://doi.org/10.3390/cmsf2023008099 - 10 Apr 2024
Abstract
Current encryption technologies mostly rely on complex algorithms or difficult mathematical problems to improve security. Therefore, it is difficult for these encryption technologies to possess both high security and high efficiency, which are two properties that people desire. Trying to solve this dilemma,
[...] Read more.
Current encryption technologies mostly rely on complex algorithms or difficult mathematical problems to improve security. Therefore, it is difficult for these encryption technologies to possess both high security and high efficiency, which are two properties that people desire. Trying to solve this dilemma, we built a new encryption technology, called configurable encryption technology (CET), based on the typical structure of reconfigurable quaternary logic operator (RQLO) that was invented in 2018. We designed the CET as a block cipher for symmetric encryption, where we use four 32-quit RQLO typical structures as the encryptor, decryptor, and two key derivation operators. Taking advantage of the reconfigurability of the RQLO typical structure, the CET can automatically reconfigure the keys and symbol substitution rules of the encryptor and decryptor after each encryption operation. We found that a chip containing about 70,000 transistors and 500 MB of nonvolatile memory could provide all the CET devices and generalized keys needed for any user’s lifetime, to implement a practical one-time pad encryption technology. We also developed a strategy to solve the current key distribution problem with prestored generalized key source data and on-site appointment codes. The CET is expected to provide a theoretical basis and core technology for using the RQLO to build a new cryptographic system with high security, fast encryption/decryption speed, and low manufacturing cost.
Full article
(This article belongs to the Proceedings of 2023 International Summit on the Study of Information)
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
iBALR3D: imBalanced-Aware Long-Range 3D Semantic Segmentation
by
Keying Zhang, Ruirui Cai, Xinqiao Wu, Jiguang Zhao and Ping Qin
Comput. Sci. Math. Forum 2024, 9(1), 6; https://doi.org/10.3390/cmsf2024009006 - 14 Mar 2024
Abstract
Three-dimensional semantic segmentation is crucial for comprehending transmission line structure and environment. This understanding forms the basis for a variety of applications, such as automatic risk assessment of line tripping caused by wildfires, wind, and thunder. However, the performance of current 3D point
[...] Read more.
Three-dimensional semantic segmentation is crucial for comprehending transmission line structure and environment. This understanding forms the basis for a variety of applications, such as automatic risk assessment of line tripping caused by wildfires, wind, and thunder. However, the performance of current 3D point cloud segmentation methods tends to degrade on imbalanced data, which negatively impacts the overall segmentation results. In this paper, we proposed an imBalanced-Aware Long-Range 3D Semantic Segmentation framework (iBALR3D) which is specifically designed for large-scale transmission line segmentation. To address the unsatisfactory performance on categories with few points, an Enhanced Imbalanced Contrastive Learning module is first proposed to improve feature discrimination between points across sampling regions by contrasting the representations with the assistance of data augmentation. A structural Adaptive Spatial Encoder is designed to capture the distinguish measures across different components. Additionally, we employ a sampling strategy to enable the model to concentrate more on regions of categories with few points. This strategy further enhances the model’s robustness in handling challenges associated with long-range and significant data imbalances. Finally, we introduce a large-scale 3D point cloud dataset (500KV3D) captured from high-voltage long-range transmission lines and evaluate iBALR3D on it. Extensive experiments demonstrate the effectiveness and superiority of our approach.
Full article
(This article belongs to the Proceedings of The 2nd AAAI Workshop on Artificial Intelligence with Biased or Scarce Data (AIBSD))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
The Exploration of High Quality Education in Scientific and Technological Innovation Based on Artificial Intelligence
by
Xiaoli Yang and Songbai Wang
Comput. Sci. Math. Forum 2023, 8(1), 98; https://doi.org/10.3390/cmsf2023008098 - 26 Feb 2024
Abstract
This paper explains that setting up artificial intelligence courses can clearly enhance students’ interest in high technology, boost learning confidence and promote students’ overall development in the following three aspects: the significance of artificial intelligence education to students, the confusion regarding artificial intelligence
[...] Read more.
This paper explains that setting up artificial intelligence courses can clearly enhance students’ interest in high technology, boost learning confidence and promote students’ overall development in the following three aspects: the significance of artificial intelligence education to students, the confusion regarding artificial intelligence teaching in this stage, especially in rural middle schools, and some related suggestions.
Full article
(This article belongs to the Proceedings of 2023 International Summit on the Study of Information)
Open AccessProceeding Paper
Exploring 3D Object Detection for Autonomous Factory Driving: Advanced Research on Handling Limited Annotations with Ground Truth Sampling Augmentation
by
Matthias Reuse, Karl Amende, Martin Simon and Bernhard Sick
Comput. Sci. Math. Forum 2024, 9(1), 5; https://doi.org/10.3390/cmsf2024009005 - 18 Feb 2024
Abstract
Autonomously driving vehicles in car factories and parking spaces can represent a competitive advantage in the logistics industry. However, the real-world application is challenging in many ways. First of all, there are no publicly available datasets for this specific task. Therefore, we equipped
[...] Read more.
Autonomously driving vehicles in car factories and parking spaces can represent a competitive advantage in the logistics industry. However, the real-world application is challenging in many ways. First of all, there are no publicly available datasets for this specific task. Therefore, we equipped two industrial production sites with up to 11 LiDAR sensors to collect and annotate our own data for infrastructural 3D object detection. These form the basis for extensive experiments. Due to the still limited amount of labeled data, the commonly used ground truth sampling augmentation is the core of research in this work. Several variations of this augmentation method are explored, revealing that in our case, the most commonly used is not necessarily the best. We show that an easy-to-create polygon can noticeably improve the detection results in this application scenario. By using these augmentation methods, it is even possible to achieve moderate detection results when only empty frames without any objects and a database with only a few labeled objects are used.
Full article
(This article belongs to the Proceedings of The 2nd AAAI Workshop on Artificial Intelligence with Biased or Scarce Data (AIBSD))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Small Dataset, Big Gains: Enhancing Reinforcement Learning by Offline Pre-Training with Model-Based Augmentation
by
Girolamo Macaluso, Alessandro Sestini and Andrew D. Bagdanov
Comput. Sci. Math. Forum 2024, 9(1), 4; https://doi.org/10.3390/cmsf2024009004 - 18 Feb 2024
Abstract
Offline reinforcement learning leverages pre-collected datasets of transitions to train policies. It can serve as an effective initialization for online algorithms, enhancing sample efficiency and speeding up convergence. However, when such datasets are limited in size and quality, offline pre-training can produce sub-optimal
[...] Read more.
Offline reinforcement learning leverages pre-collected datasets of transitions to train policies. It can serve as an effective initialization for online algorithms, enhancing sample efficiency and speeding up convergence. However, when such datasets are limited in size and quality, offline pre-training can produce sub-optimal policies and lead to a degraded online reinforcement learning performance. In this paper, we propose a model-based data augmentation strategy to maximize the benefits of offline reinforcement learning pre-training and reduce the scale of data needed to be effective. Our approach leverages a world model of the environment trained on the offline dataset to augment states during offline pre-training. We evaluate our approach on a variety of MuJoCo robotic tasks, and our results show that it can jumpstart online fine-tuning and substantially reduce—in some cases by an order of magnitude—the required number of environment interactions.
Full article
(This article belongs to the Proceedings of The 2nd AAAI Workshop on Artificial Intelligence with Biased or Scarce Data (AIBSD))
►▼
Show Figures

Figure 1
Open AccessProceeding Paper
Study on the Application of Virtual Reality Technology in Cross-Border Higher Education
by
Yanfang Hou
Comput. Sci. Math. Forum 2023, 8(1), 97; https://doi.org/10.3390/cmsf2023008097 - 8 Feb 2024
Abstract
This paper summarizes the problems existing in cross-border higher education through the analysis of the development status and characteristics of virtual reality technology and cross-border higher education, and puts forward the important significance and enlightenment of the application of virtual reality technology in
[...] Read more.
This paper summarizes the problems existing in cross-border higher education through the analysis of the development status and characteristics of virtual reality technology and cross-border higher education, and puts forward the important significance and enlightenment of the application of virtual reality technology in cross-border higher education in the new era for solving the practical problems of cross-border education. It also points out that the new mode and situation of online and offline joint development created by the integration of virtual reality technology and cross-border higher education will have an important impact on accelerating the opening up of Chinese education and improving the quality and efficiency of cross-border higher education.
Full article
(This article belongs to the Proceedings of 2023 International Summit on the Study of Information)
Highly Accessed Articles
Latest Books
E-Mail Alert
News
Topics
