Virginia Tech Innovation Campus’ Post

📄 On 08th October 2024, my paper titled '𝘐𝘮𝘢𝘨𝘦 𝘚𝘦𝘨𝘮𝘦𝘯𝘵𝘢𝘵𝘪𝘰𝘯 𝘚𝘶𝘳𝘷𝘦𝘺: 𝘊𝘭𝘢𝘴𝘴𝘪𝘤𝘢𝘭 𝘢𝘯𝘥 𝘋𝘦𝘦𝘱 𝘓𝘦𝘢𝘳𝘯𝘪𝘯𝘨 𝘔𝘦𝘵𝘩𝘰𝘥𝘴' was published on IEEE Xplore. This paper was presented at the 4ᵗʰ International Conference on Electrical, Computer & Energy Technologies (ICECET) in Sydney, Australia on 25th July 2024. ⌛ 8 months ago, I had taken what I would consider one of my greatest risks; to write a survey paper with: - 0 knowledge on most Computer Vision techniques. - 0 knowledge on how to structure a research paper, and with. - 15 days to submit the final version. Regardless of the time crunch, here are the milestones to its success: 🗓️ 16th March 2024 Research begins, dedicating 10 hours a day into completing it within 2 weeks. This includes research into Computer Vision, surveying segmentation techniques, and collating performances across 100+ models from 30+ authors (and understanding LaTeX). 🗓️ 03rd July 2024 The paper is accepted and will be presented by me at ICECET '24 in Sydney, Australia on 25th July 2024! Read more on the learning journey: https://lnkd.in/gWsNSgKS 🗓️ 25th July 2024 I had the immense opportunity to not only present my paper but to meet like-minded individuals sharing the same goal as me - to incorporate field knowledge and share our findings with the world! Read more on takeaways from the conference: https://lnkd.in/gAHQNq8z 👏 Hats off to my co-authors Abhijit Dasgupta, Hazem El-Alfy, and Dr. Anish Roychowdhury, Ph.D. for playing pivotal roles in the success of this 8-month journey. 🚀 This paper is now yours to read https://lnkd.in/g7mcvRQM Onwards & Upwards, Viren Bachani Reflection_010 #icecet #conference #computervision #sydney #australia

📝 Recent article on cyber-security from our team 🧖♂️🧖♂️. Feel free to read 📖 . The article comprehensively reviews email spam detection methods 📩 , emphasizing machine learning and deep learning techniques. The paper evaluates various detection methods, the effectiveness of machine learning and deep learning techniques, challenges, and future research opportunities, aiming to improve spam email filtering accuracy. Read on -> https://lnkd.in/g6whxZbt

Welcome to read and share the newly published paper "The Novel EfficientNet Architecture-Based System and Algorithm to Predict Complex Human Emotions ". This excellent paper is written by Mavlonbek Khomidov and Jong-Ha Lee Read via: https://lnkd.in/g4xSgbpe MDPI Keimyung University #biosignal #facialexpression #emotion #algorithms

Excited to share a new publication! We are very much interested in the lab in exploring the potential benefits of combining iterative optimization with deep learning. In our recent paper lead by Yoav Noah, which was just accepted for publication in the IEEE Transactions on Mobile Computing, we showed that in the context of distributed optimization, integrating learning can greatly facilitate approaching consensus within a limited number of communication rounds (or at least getting the best performance in terms of the available data). The full paper is already available as early access at https://lnkd.in/d86C9nhS

On how to characterize the duration and amplitude of unwanted events and semantically describe them applied to wireless links. Joint work with Blaž Bertalanič and Vid Hanžel. #researchpaper #artificialintelligence #digitaltwin #creativity

📢 Discover valuable insights on managing technical debt in embedded systems! 📘 Delve into "The Perception of Technical Debt in the Embedded Systems Domain: An Industrial Case Study," a compelling research paper by Ampatzoglou, Chatzigeorgiou, Avgeriou, Abrahamsson, Martini, Zdun, and Systä, presented at the 2016 IEEE 8th International Workshop on Managing Technical Debt (MTD). This study is a must-read for professionals and researchers seeking to understand the crucial impact of technical debt on system sustainability and software quality. Gain practical, evidence-based knowledge that can transform your strategies for managing technical debt, enhancing long-term project outcomes. To explore the findings and implications of this groundbreaking research, visit the full article on GPT-Lab's website: [Read more](https://lnkd.in/dRAfgjrP). We welcome your thoughts and encourage you to share with your network! This post is created and published by our AI helpers. #TechnicalDebt #EmbeddedSystems #SoftwareEngineering #IEEE #ResearchInsights #IndustryTransformation #AIAssistedWriting

📣 A Survey on Goal-Oriented Semantic Communication: Techniques, Challenges, and Future Directions [1]. 🎯 Because semantic communication (SemCom) and goal-oriented SemCom lead to minimum power usage, bandwidth consumption, and transmission delay, they are promising 6G technology enablers — i.e., SemCom and goal-oriented SemCom for 6G [1], [2]. 🎯 On the other hand, 6G is essential for realizing the various use cases of SemCom and goal-oriented SemCom — i.e., 6G for SemCom and goal-oriented SemCom [1], [2]. 🎯 The mentioned paradigms of (goal-oriented) SemCom for 6G and 6G for (goal-oriented) SemCom call for a tighter integration of 6G, SemCom, and goal-oriented SemCom [1], [2]. 📚 The IEEE Access recently published the early-access version of our comprehensive survey on goal-oriented SemCom [1], which has also exposed the fundamental challenges of 6G. 🔎 A joint work with Prof. Drs. Georges Kaddoum and Mehdi Bennis, my published survey has delivered the following contributions: 1️⃣ It first discusses the fundamental challenges of 6G. 2️⃣ It then provides a taxonomy for the state-of-the-art goal-oriented SemCom research. 3️⃣ It then details the state-of-the-art research landscape of goal-oriented SemCom. 4️⃣ It then presents the major state-of-the-art trends and use cases of goal-oriented SemCom. 5️⃣ It then outlines state-of-the-art mathematical frameworks of goal-oriebted SemCom. 6️⃣ It then discusses state-of-the-art theories of goal-oriented SemCom. 7️⃣ It then exposes the fundamental and major challenges of goal-oriented SemCom theories, algorithms, and realization. 8️⃣ Finally, it offers promising future directions for goal-oriented theories, algorithms, and realization. 🌟 By exposing the 6G fundamental challenges; revealing the fundamental and major challenges of goal-oriented SemCom; and offering novel future directions for goal-oriented SemCom theories, algorithms, and realization, the survey in [1] duly stimulates many lines of research on goal-oriented SemCom theories, algorithms, and realization. 🤝 I acknowledge Prof. Georges Kaddoum and Prof. Mehdi Bennis for their contributions and thoughtful leadership, and Dr. Hamid Gharavi for funding and leadership support. 🤝 I acknowledge the U.S. Department of Commerce and its agency the U.S. National Institute of Standards and Technology (NIST) for partially funding this work. 🤝 I acknowledge — on behalf of Prof. Georges Kaddoum — the Natural Sciences and Engineering Research Council of Canada (NSERC) and its Canada Research Chairs Program for the funding support of this work. 🤝 I acknowledge the more than 1000 authors, cited by our survey, for their insightful contributions that have informed our survey. [1] https://lnkd.in/eeYJwE4H [2] https://lnkd.in/eVGYJEyv #6g #wirelesscommunication #ai #artificialintelligence #deeplearning #wirelesscommunications

AI Takes Center Stage in China's First "Hospital Town" We round off the week with a groundbreaking hybrid intelligence case from China. State media has announced the country's first AI hospital town, a revolutionary concept where virtual patients are treated by AI doctors. Developed by Tsinghua University, this system aims to revolutionize medical consultations by training doctor agents in a simulated environment. This allows them to evolve independently and refine their disease treatment skills. The potential impact is significant: researchers believe this model can pave the way for real-world applications of AI doctors, ultimately leading to high-quality, affordable, and convenient healthcare for everyone. The "Agent Hospital" goes beyond just diagnosis. It also allows real doctors to interact with virtual patients, providing an invaluable training ground for medical students. Students can develop treatment plans for a wide range of simulated patients, gaining valuable experience without risk to real patients. This virtual world is powered by large language models (LLMs), creating intelligent agents for doctors, nurses, and patients, enabling them to interact autonomously within the simulated environment. The Agent Hospital's AI doctor agents are showing promise, achieving a 93.06% accuracy rate on a medical licensing exam for respiratory diseases. https://lnkd.in/gk8_QAEG #AI #hybridintelligence #Healthcare #Innovation #China #TsinghuaUniversity

With great pleasure, I would like to share the publication of a book on semantic communications, I co-edited along with Christina Chaccour, Walid Saad and Merouane Debbah. It was a great pleasure to work with many esteemed authors across the globe to edit a first of its kind (as far as we know) book on semantic communications. Without the help of their contributions, this would not have been a possibility. https://lnkd.in/ejDHTF8M I invite all researchers in academia and industry who are interested in next-generation networking (6G and beyond) to read this book and share your feedback or opinions. In particular, this book delves into diverse aspects such as: a) a unique perspective on semantic information, b) networking fundamentals of semantic communication systems, c) machine reasoning perspective on semantic communications, and d) security challenges of semantic communications.

🎉 Happy to share that the final version of our paper with Bünyamin Kartal and Aykut Koç on generalizing the fractional Fourier transform (FRFT) to the graph domain, titled "Graph Fractional Fourier Transform: A Unified Theory," published in IEEE Transactions on Signal Processing, is now available in IEEE Xplore! ✨ Highlights of our contributions include: 🔎 A rigorous extension of the fractional power-based definition of GFRFT to support any graph structure and transform order. 🔎 Introduction of a novel hyper-differential operator-based GFRFT definition, ensuring consistency with classical FRFT theory and enabling faster computations for large graphs. 🔎 A procedure to optimally select the transform order via learning, enabling effective application of GFRFT within neural networks and machine learning applications. 🔎 Comprehensive experimental validation demonstrating the equivalence of GFRFT definitions and the advantages over GFT and other GSP methods in tasks such as denoising, classification, and sampling. We believe this unified theory for GFRFT will open up new research venues and provide a powerful tool for analyzing graph-based data. Check out the paper for more details! 📰 Paper: https://lnkd.in/dVnM_wqk ⚙ Code: https://lnkd.in/d6zspwD9 For further information on using the FRFT in joint time-vertex signal processing and intrinsic of trainable transform order, please see our related works respectively: - T. Alikaşifoğlu, B. Kartal and A. Koç, "Wiener Filtering in Joint Time-Vertex Fractional Fourier Domains," in IEEE Signal Processing Letters, vol. 31, pp. 1319-1323, 2024, doi: 10.1109/LSP.2024.3396664. https://lnkd.in/dbfUSXvC - E. Koç, T. Alikaşifoğlu, A. C. Aras and A. Koç, "Trainable Fractional Fourier Transform," in IEEE Signal Processing Letters, vol. 31, pp. 751-755, 2024, doi: 10.1109/LSP.2024.3372779. https://lnkd.in/dkuEUaSj #Graph #SignalProcessing #MachineLearning #Fourier