Invited Speakers
2024 9th International Conference on Intelligent Information Technology (ICIIT 2024) aims to gather professors, researchers, scholars and industrial pioneers all over the world. ICIIT is the premier forum for the presentation and exchange of past experiences and new advances and research results in the field of theoretical and industrial experience. The conference welcomes contributions which promote the exchange of ideas and rational discourse between educators and researchers all over the world. We aim to building an idea-trading platform for the purpose of encouraging researcher participating in this event. ICIIT 2024 is welcome qualified persons to delivery a speech in the related fields. If you are interested, please send a brief CV with photo to the conference email box: iciit@cbees.net.
Invited Speaker I
Assist. Prof. Balachandran Manavalan
Sungkyunkwan University (SKKU), South Korea
Dr. Balachandran Manavalan is an Assistant Professor at the Department of Integrative Biotechnology, Sungkyunkwan University (SKKU), South Korea. He obtained his Ph.D. in Computational Biology from Ajou University in 2011. He spent 10 years honing his research expertise, serving as a research fellow at the KIAS, and a research assistant professor at Ajou University School of Medicine, South Korea. He established his own research group at SKKU’s Department of Integrative biotechnology in 2022. His research interests include artificial intelligence, bioinformatics, machine learning, big data, proteomics, and functional genomics. He has been developing cutting-edge bioinformatics tools for identifying peptide therapeutic functions, post-transcriptional modifications, DNA epigenetic modification sites, and RNA post-transcriptional modifications. His impressive publication record includes over 100 papers, the majority of which are published in top-tier journals ranked within the top 10% of the Journal Citation Reports (JCR). His remarkable achievements have earned him recognition as a top 2% highly cited researcher for the past four consecutive years, according to Stanford University data.
Speech Title: "Artificial Intelligence to Explore Multimodality Data and Accelerate Biomedical Knowledge Discovery"
Abstract: This talk introduces two of our recent AI-driven methodologies: MeL-STPhos for identifying SARS-CoV2 phosphorylation sites and H2Opred for identifying 2’-O-methylation (2OM) sites in human RNA. Firstly, MeL-STPhos, a meta-learning model, for the accurate prediction of SARS-Cov2 phosphorylation sites. Briefly, we conducted a comprehensive exploration of 29 feature descriptors, assessed each descriptor’s ability using 14 distinct classifiers, and identified the best-performing model for each descriptor. These individual models were then synergistically combined into a robust meta-model. Notably, we developed both cell-specific and generic models and demonstrated their practical application scenarios. Secondly, the novel hybrid deep learning model, H2Opred, for the identification of 2OM sites in human RNA. This is the first method that showed an advantage of having a generic model compared to the nucleotide-specific models. H2Opred integrates stacked 1D convolutional neural network and attention-based bidirectional gated recurrent units blocks to learn and extract multi-modal features, both conventional descriptors and NLP-based embeddings. This approach markedly improves predictive accuracy over conventional ML-based models and state-of-the-art methods. These AI-driven methodologies not only showcase the potential of AI in unraveling intricate cellular processes but also accelerate biomedical knowledge discovery.
Invited Speaker II
Dr. Hung Dang
FPT Corporation, Vietnam
Dr. Hung Dang is currently the Head of Research in the FPT CTO's office, and has been lecturing at the Vietnam National University. Prior to joining FPT, he was a Research Fellow at the Singapore National Research Foundation’s Strategic Capability Research Centre in Privacy-preserving Technologies. His research focuses on computer security and distributed systems, and has received numerous scientific citations.
Speech Title: "What Else Does AI Stand For"
Abstract: Large Language Models and more recently Large Vision Models have emerged spectacularly and left the world in awe. They are commonly referred to as instances of Artificial Intelligence, or AI for short. Much has been discussed about their applications, their potentials, and even what existential threats they could be. Nonetheless, it remains unclear to what extent those discussed points are realistically imminent, and which parts of them are purely hype-driven. This talk will take a more sober approach to the notion of AI, attempting to suggest different connotations of AI.
Invited Speaker III
Prof. Masaomi Kimura
Shibaura Institute Technology, Japan
Prof. Masaomi Kimura received a Ph.D. degree from The University of Tokyo. After his career as a system engineer in IBM, he started his career as a researcher at Shibaura Institute of Technology (SIT) in 2004. Now, he is a director of the international exchange center from 2023. He is also a professor in the Department of Computer Science and Engineering in the College of Engineering. His current research interests are in the areas of data science and data engineering, with a particular focus on data analysis as an application of artificial intelligence (machine learning), especially using deep learning. His research ranges from developing novel models for data analysis to their application in solving real-world problems such as automatic generation of ASCII art for transmission data compression and adversarial attack and defense methods to improve security of deep learning models.
Speech Title: "Low Visibility Perturbation in Adversarial Attacks on Image Recognition"
Abstract: Deep learning neural networks are a promising solution for solving complex real-world problems. They are particularly important in image recognition and are being applied to new technologies such as automated driving. However, adversarial examples, which apply perturbations to image pixels, pose a threat to the safety of deep learning neural networks. This presentation will provide an overview of the concept of adversarial examples and introduce a technique that makes it difficult to visually distinguish when a perturbation has been added.