Loading…
Type: Virtual Room 4D clear filter
Friday, May 23
 

8:58am EDT

Opening Remarks
Friday May 23, 2025 8:58am - 9:00am EDT
Invited Guest/Session Chair
avatar for Prof. Tanupriya Choudhury

Prof. Tanupriya Choudhury

Professor, University of Petroleum and Energy Studies, India
avatar for Prof. Vishnu Kumar

Prof. Vishnu Kumar

Assistant Professor, Morgan State University, United States
Friday May 23, 2025 8:58am - 9:00am EDT
Virtual Room D New York, USA

9:00am EDT

A Soft Actor-Critic approach for Energy Management in Microgrids
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Monica Alonso, Hortensia Amaris, Maria Angeles Moreno, Farzaneh Abdollahi, Lucia Gauchia
Abstract - This paper introduces a novel approach to applying artificial intelligence algorithms based on Reinforcement Learning (RL) for microgrid energy management. Two energy storage systems are considered: stationary battery storage and electric vehicle batteries with G2V/V2G capability. The proposed energy management algorithm considers (i) the uncertainty of photovoltaic energy production, (ii) fluctuations in electricity market prices, and (iii) driver anxiety concerning the vehicle’ range at departure time. The significance of specific parameters, such as time horizon selection and the constant value related to the electric vehicle driver’s anxiety, are examined to optimise the RL reward. Results demonstrate the algorithm's excellent performance under different scenarios.
Paper Presenter
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

Achieving Optimal Tissue Repair Through MARL with Reward Shaping and Curriculum Learning
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Muhammad Al-Zafar Khan, Jamal Al-Karaki, Marwan Omar
Abstract - In this paper, we present a multi-agent reinforcement learning (MARL) framework for optimizing tissue repair processes using engineered biological agents. Our approach integrates: (1) Stochastic reaction-diffusion systems modeling molecular signaling, (2) Neural-like electrochemical communication with Hebbian plasticity, and (3) A biologically informed reward function combining chemical gradient tracking, neural synchronization, and robust penalties. A curriculum learning scheme guides the agent through progressively complex repair scenarios. In silico experiments demonstrate emergent repair strategies, including dynamic secretion control and spatial coordination.
Paper Presenter
avatar for Muhammad Al-Zafar Khan

Muhammad Al-Zafar Khan

United Arab Emirates
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

Enhancing User Adoption of Digital Fuel Subsidy Systems: Insights from the MyPertamina App Using UTAUT Model
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Arinta Athaya Kayana, Candiwan Candiwan
Abstract - Rapid digital transformation in various sectors has changed how individuals interact with technology, including in essential services such as fuel purchases. While the MyPertamina app promises to revolutionize access to subsidized fuel, its adoption has fallen short of expectations, raising curiosity about the underlying factors influencing user uptake. This study aims to explore the factors that influence the adoption of the MyPertamina application using the Unified Theory of Acceptance and Use of Technology (UTAUT) framework. The methodology used is a quantitative approach through distributing questionnaires to MyPertamina users. The analysis results showed a significant relationship between performance expectations, government regulations, and the reliability of institutional sources on behavioral intentions and adoption. However, factors such as effort expectancy, awareness, perceived financial cost, technological infrastructure support, government regulation and institutional privacy concerns did not show a significant impact towards behavioral intention and adoption. The findings highlight the importance of improving user perceptions of the app’s performance and ensuring the reliability of institutional sources for increase behavioral intention and adoption. Additionally, refining regulatory and enhancing app features for a smoother refueling experience are crucial in encouraging broader usage. By addressing these factors, stakeholders can implement more effective strategies to increase the adoption.
Paper Presenter
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

From Structure to Experience: Designing the Front End of a Cultural Culinary Portal
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Gloria Virginia, Maria Amanda, Budi Susanto, Umi Proboyekti
Abstract - The digital preservation of traditional culinary knowledge is essential to sustaining cultural identity amid rapid technological change. This study focuses on improving the usability of a semantic web-based portal for Indonesian traditional food through a user-centered design approach. The redesign involved card sorting, tree testing, and prototype usability evaluation. Key improvements include enhanced navigation structures, clearer labeling, and the application of interface patterns such as fat menus, breadcrumbs, and carousels. Usability testing showed notable increases in success rates and navigation efficiency, with a System Usability Scale (SUS) score of 85.5, indicating excellent usability. The integration of information architecture principles significantly improved user interaction and content discoverability. This research demonstrates that combining cognitive-oriented design with semantic data models creates more intuitive and accessible cultural heritage platforms.
Paper Presenter
avatar for Gloria Virginia
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

Human-AI Collaboration for Rescue Missions: A Reinforcement Learning Approach in VR Environments
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Shahin Doroudian, Mohsen Dorodchi
Abstract - Firefighting operations in hazardous environments demand agile and adaptive strategies to effectively combat fires while safeguarding the lives of firefighters and civilians. Traditional approaches often rely on predetermined protocols, which may not adapt well to dynamic and unpredictable situations. This paper proposes a novel framework utilizing Reinforcement Learning (RL) to optimize rescue mission strategies. By harnessing the power of RL, this approach enables AI agents to learn and adapt their behavior based on environmental cues and feedback, leading to more effective and responsive rescue operations. The framework integrates various factors such as terrain and path complexity and the presence of hazards, like fire and smoke, into the decision-making process. Through iterative learning, agents evolve their strategies, identifying optimal paths and rescue tactics. Furthermore, this proposal discusses the potential benefits of employing DRL in rescue missions, including enhanced adaptability, scalability, and robustness in diverse and challenging environments. The adoption of RL to optimize strategies for rescue missions represents a significant opportunity to progress in the disaster response domain. At the end, our results show that the RL-driven method enhances rescue operation outcomes and safety by reducing rescue times, increasing the number of people rescued, optimizing resource utilization, and boosting overall efficiency.
Paper Presenter
avatar for Shahin Doroudian

Shahin Doroudian

United States of America
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

Improving Alzheimer's Disease and Related Dementias Phenotyping in Electronic Medical Recodes Through Transfer Learning
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Yijun Shao, Ying Yin, Debby Tsuang, Phillip Ma, Edward Zamrini, Ali Ahmed, Charles Faselis, Katherine Wilson, Karl Brown, Qing Zeng-Treitler
Abstract - This study involved the development and evaluation of a novel Deep Neural Network (DNN) model for Alzheimer's disease and related dementias (ADRD) phenotyping. The model was initially trained on a large cohort of 100,000 cases and controls and subsequently fine-tuned using a smaller, expert reviewed dataset of 1,200 individuals. The final fine-tuned model achieved an Area Under the Receiver Operating Characteristic curve (AUC) of 0.832. For further validation, the model's predictive capability was assessed in a separate randomly selected patient cohort comprising individuals without an ADRD diagnosis from 2009 to 2018. The survival analysis shows that patients with higher predicted ADRD risk scores exhibited a significantly increased incidence of developing ADRD after their index date within five years.
Paper Presenter
avatar for Yijun Shao

Yijun Shao

United States of America
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

Key Challenges to Develop Agentic News AI on Flutter
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Bat-Erdene Batsukh, Ariunbold Tsoodol
Abstract - Agentic AI significantly enhance the efficiency and quality of news writing by generating engaging headlines and concise summaries while maintaining the accuracy and objectivity of the original content. These tools leverage advanced algorithms and large language models (LLMs) to streamline the content creation process, ultimately improving audience engagement and journalistic productivity. The approaches range from using neural network algorithms to LLMs and question-answering systems, each with its unique strengths and challenges. Agent, like News writers, operate in a dynamic and fast-paced environment where staying updated with current global events is crucial. The digital era has transformed how journalists gather, verify, and disseminate information. The integration of Natural Language Processing (NLP) into a Flutter-based news app offers an amazing way to deliver personalized news feeds tailored to individual user experiences. By leveraging NLP techniques, the app can analyse user behaviour, understand content, provide relevant news recommendations, neural machine translations and news content summarizations. Developing a centralized AI-powered news app using Flutter presents a unique set of challenges that span technical, ethical, and operational domains. We conducted this research to discover ways to overcome these challenges and spectacles. In particular, we used artificial intelligence to automatically translate news into the reader's native language and to provide a summary view instead of a large amount of text when reading in detail.
Paper Presenter
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

9:00am EDT

Mass-parallel Sleptsov Net-Based Solving PDEs on FPGA for Embedded Control
Friday May 23, 2025 9:00am - 11:00am EDT
Authors - Dmitry A. Zaitsev, Alistair A. McEwan, Alexander A. Kostikov
Abstract - Real-time embedded applications are normally viewed as continuous processes and are often specified using Partial Differential Equations (PDEs) and with certain boundary conditions. In this paper we present techniques for fast mass-parallel numerical solving of PDEs. We compose specialized lattices based on the integer number approximation specified with Sleptsov nets to be implemented as dedicated hardware, which we prototype on an FPGA. For mass-parallel solving of PDEs, we employ ad-hoc finite-difference schemes and iteration methods that allow us to recalculate the lattice values in a single time cycle with appropriate accuracy suitable for control of hypersonic objects and thermonuclear reactions.
Paper Presenter
avatar for Dmitry A. Zaitsev

Dmitry A. Zaitsev

United Kingdom
Friday May 23, 2025 9:00am - 11:00am EDT
Virtual Room D New York, USA

11:00am EDT

Session Chair Concluding Remarks
Friday May 23, 2025 11:00am - 11:02am EDT
Invited Guest/Session Chair
avatar for Prof. Tanupriya Choudhury

Prof. Tanupriya Choudhury

Professor, University of Petroleum and Energy Studies, India
avatar for Prof. Vishnu Kumar

Prof. Vishnu Kumar

Assistant Professor, Morgan State University, United States
Friday May 23, 2025 11:00am - 11:02am EDT
Virtual Room D New York, USA

11:02am EDT

Session Closing and Information To Authors
Friday May 23, 2025 11:02am - 11:05am EDT
Moderator
Friday May 23, 2025 11:02am - 11:05am EDT
Virtual Room D New York, USA
 
Share Modal

Share this link via

Or copy link

Filter sessions
Apply filters to sessions.