The 5th International Conference on Signal Processing and Machine Learning (CONF-SPML 2025) was a hybrid conference which includes several symposium series (offline and online) around the world. Dr. Stavros Shiaeles from University of Portsmouth, Prof. Anil Fernando from University of Strathclyde, Dr. Bilyaminu Romo Auwal from University of East London, and Dr. Xinqing Xiao from China Agricultural University have chaired these symposium series on related topics. CONF-SPML 2025 provided the participants with good opportunities to exchange ideas and build networks, and it will lead to further collaborations between both universities and other societies.
Symposium Chair: Dr. Stavros Shiaeles, Associate Professor in University of Portsmouth
The advent of deepfake technology, driven by advancements in artificial intelligence, has revolutionized how digital media can be manipulated. Using techniques like generative adversarial networks (GANs), deepfakes can convincingly alter images, videos, and audio, making it increasingly difficult to distinguish between authentic and fabricated content. As these technologies become more sophisticated and accessible, they pose significant challenges to the integrity of media across various domains, including journalism, law enforcement, and personal privacy. Understanding and detecting deepfakes is critical in maintaining trust in digital media and protecting against the spread of misinformation.
Symposium Chair: Prof. Anil Fernando, Professor in University of Strathclyde
The workshop focuses on the semantic security of autoencoder-based communication systems in the context of image transmission. We will demonstrate the effectiveness of inherent security in semantic communications using the MNIST dataset of handwritten digits. Two separate autoencoders are trained on different subsets of the MNIST dataset, simulating different scenarios where a legitimate encoder and an attacker have access to distinct autoencoder models trained on the same data distribution. The methodology involves training the autoencoders, obtaining latent representations from one autoencoder's encoder, and attempting to reconstruct the original images using both the matching decoder and the mismatched decoder from the other autoencoder. The reconstructed images are then analyzed and compared to assess the level of semantic security achieved by the mismatch between the encoder and decoder models. Participants will get hands on experience how semantic communication test bed can be used to secure image transmissions.
Symposium Chair: Dr. Bilyaminu Romo Auwal, Senior Lecturer in University of East London
The recent workshop "The Digital Economy: Harnessing Financial Innovations for a Sustainable Future" brought together experts across blockchain, fintech, cloud computing, and IoT to explore transformative solutions for sustainability reporting and regenerative finance (ReFi). As chair, I emphasized how ReFi principles can reshape our approach to environmental impact tracking and sustainable development. The keynote address highlighted the convergence of digital technologies in driving sustainable economic transformation. Participants explored how blockchain's immutable ledger technology, combined with IoT sensors and cloud computing infrastructure, can create transparent and verifiable sustainability reporting systems. This integration addresses the critical challenge of greenwashing while supporting progress toward UN Sustainable Development Goals. A significant focus was placed on standardizing data collection methods through IoT sensors and leveraging blockchain for data integrity. The discussion emphasized how ReFi mechanisms can incentivize positive environmental impact through innovative financial instruments and transparent reporting frameworks. Cloud computing emerged as a crucial enabler, providing the necessary computational power for processing complex environmental datasets. The workshop concluded with practical recommendations for implementing blockchain-based platforms for sustainability verification and exploring fintech solutions for green financing. Participants agreed that the intersection of these technologies, coupled with ReFi principles, presents a promising pathway toward achieving global net zero emissions goals while ensuring economic prosperity. Moving forward, the workshop established a foundation for continued collaboration between technologists, sustainability experts, and financial innovators in building a more regenerative and transparent digital economy.
Symposium Chair: Dr. Xinqing Xiao, Associate Professor in China Agricultural University
Hybrid intelligent systems could hybrid many advanced information technologies together for achieving the high intelligent applications. The advances information technologies include the hardware and software system, such as the intelligent sensors, computer software, machine learning, signal processing, database, decision making systems etc. They have been appeared and applied in many areas such as intelligent industry, intelligent agriculture and intelligent building recently. Here, the workshop will examine some of the recent publications in leading journals that cover hybrid intelligent systems topics or have implications for hybrid intelligent systems and applications. We will also introduce some hybrid intelligent systems technologies for mainly in intelligent agriculture such as food processing. The scholars (including doctoral and master students) from all disciplines and practitioners interested in the hybrid intelligent systems and applications are all encouraged to attend.
The online session of the International Conference on Signal Processing and Machine Learning (CONF-SPML 2025) was held on January 15, 2025. Dr. Stavros Shiaeles from the University of Portsmouth, Prof. Anil Fernando from University of Strathclyde, Dr. Bilyaminu Romo Auwal from University of East London, and Dr. Xinqing Xiao from China Agricultural University have given keynote speeches on related topics of machine learning, deep learning, etc.
Accepted papers of CONF-SPML 2025 have been published in IET Conference Proceedings (Online ISSN: 2732-4494) or Applied and Computational Engineering (Print ISSN 2755-2721), and have been submitted to EI Compendex, Conference Proceedings Citation Index (CPCI), Crossref, Portico, Inspec, Google Scholar, CNKI, and other databases for indexing.
Title: IET Conference Proceedings
Press: IET
ISSN: Online ISSN: 2732-4494