2024 ICONIP Special Sessions

We are pleased to advise that the following Special Sessions are confirmed for the 2024 programme and we would like to thank the organisers for the time they took to put forward and manage these sessions.

No.Special Session TitleRuntime (hours)Special Session Organiser
1Advances in Deep Learning for Biometrics and Its Applications4Larbi Boubchir
2AI and Game Production2Samah Hassan Abd El Maksoud
3Trends in Swarm Intelligence Optimization Assisted by Machine Learning Techniques2Xinyu Zhou
4Recent Advances in AI-empowered Oceanic Computing2Ruobin Gao
5Engineering Applications of Hybrid Artificial Intelligence Tools4Zbigniew Gomolka
6Computational Cognitive Neuroscience4Zohreh Doborjeh
7Reliable, Robus and Secure Machine Learning Algorithms4Harry Nguyen
8AI in Environmental, Conservation and Geospatial Applications4Akbar Ghobakhlou
9AI Education2Michael Watts
10Neural Models of Infants and Child Development4Alistair Knott
11Computer Vision and Sensor Signal Processing for Enhancing Life Quality and Safety4Boris Bačić

1. Advances in Deep Learning for Biometrics and Its Applications

Led by Prof. Larbi Boubchir, Prof. Boubaker Daachi (University of Paris 8)

The biometric is a growing technology due to the needs of the society, companies and governments for recognition, security and privacy concerns. It has also become a growing research area that offers greater security and convenience solutions for various applications in biometrics.

The advances in artificial intelligence, in particular in artificial learning, have allowed to resolve various complex problems related to recognition, detection, control, security, forensic identification, etc. Indeed, most of biometric technologies are based on a typical representation, including biometric data processing, feature extraction, and classification process. Deep learning offers an end-to-end learning paradigm allowing to unify these parts. It has been shown to be a promising and powerful alternative to conventional approaches based on machine learning.

This special session aims to bring together researchers, scientists and industry professionals to present and discuss their recent contributions in deep learning for biometrics and its applications.

The main topics that are of interest to this special session include, but are not limited to, the following:

  • Deep learning for biometrics authentication and identification
  • Physiological and behavioral biometrics for recognition and security (e.g., fingerprint, palmprint, palm vein, face, iris, ear, gait, voice, etc.)
  • Soft biometrics
  • Multimodal biometrics
  • Big Data challenges in biometrics
  • Attacks on biometric systems
  • Security and privacy in biometrics
  • Forensic identification
  • Emerging biometrics
  • Related applications

2. AI and Game Production

Led by Dr. Samah Hassan Abd El Maksoud, Dr. Aslıhan Tece Bayrak (Media Design School)

Artificial Intelligence is an integral element of game development, where AI has been an effective tool for compelling gameplay experiences such as the use of non-player characters or intelligent agents to enhance player experience and engagement by offering adaptive and/or responsive gameplay. The introduction of AI as a development tool to enhance the game production process is a fascinating prospect given how pivotal AI has become in diverse fields including but not limited to engineering, design, hospitality, marketing, or creative technologies. Although still in early days of adoption, the application of AI in game production spans from creating content, aiding in design and coding, planning, testing, analysing behaviour and even marketing. While the broad use of AI in games highlights the versatility of AI in addressing the creative and technical challenges of game production, it presents challenges such as ensuring ethical use of AI, and protecting data privacy and intellectual property,

Topics of interest include but they are not limited to:

  • The integration of generative AI in game production pipelines.
  • AI as a game design assistant.
  • AI as an art apprentice.
  • AI and the SDLC of games.
  • AI and game marketing and player acquisition.
  • Emotion AI-Affective computing and games.
  • Ethics and implications of the use of AI in game development/production.

We welcome contributions on explorations for the use of AI as a tool in any stage of game production. Some examples could be:

  • Case studies on development and/or integration of an AI tool into specific areas of game production.
  • Technical studies that focus on production and/or integration guidelines for specialised AI tools.
  • Training and/or fine-tuning strategies for language models.
  • Analysis or position papers discussing aspects of intellectual property concerning game development know-how, data privacy of companies, and implications of integrating AI into game production process.

This session includes a brief presentation by an invited speaker followed by a paper session to present the accepted manuscripts.

3. New Trends of Swarm Intelligence Optimization Assisted by Machine Learning Techniques

Led by Dr. Xinyu Zhou, Dr. Wenlong Ni (Jiangxi Normal University), Dr. Hu Peng (Jiujiang University), Prof. Hui Wang (Nanchang Institute of Technology), Dr. Wei Li (Jiangxi University of Science and Technology)

As an appealing optimization methodology, Swarm Intelligence Optimization (SIO) has been shown impressive advantages in dealing with complex real-world optimization problems, including continue problems and combinatorial problems, such as the neural architecture search (NAS) and feature selection for a learning algorithm. In recent years, to pursue higher efficiency and effectiveness, there has been a fair amount of research into the design of SIO using machine learning techniques. For example, the reinforcement learning is embedded into the SIO procedure for selecting search strategies or adjusting hyperparameters in solving robot path planning problem, parameters extraction of photovoltaic models, and sewage treatment control problem. Similar examples can be easily given, such as the SVM-based SIO to solve computationally expensive problems. Due to its attracting performance, this session aims to investigate in both the new theories and methods on SIO assisted by machine learning techniques, and the applications in real-world problems. The submissions on various aspects of algorithm design and applications will be welcomed, such as the assistances from neighborhood learning technique, fitness landscape analysis, data-driven technique, PCA, and reinforcement learning.

Topics include (but are not restricted to):

  1.  Reinforcement learning based SIO
  2. Dynamic neighborhood learning
  3. Online fitness landscape analysis technique
  4. Large scale NAS optimization
  5. Neuroevolution
  6. Multi-swarm and self-adaptive approaches
  7. SIO in dynamic or uncertain environment
  8. Combinations with local search techniques
  9. Multi-task optimization
  10. Multi-objective optimization
  11. Mixed variable optimization

4. Recent advances in AI-empowered oceanic computing

Led by Dr. Ruobin Gao (Nanyang Technological University), Prof. P. N. Suganthan (Qatar University)

Oceanic computing is vital for harnessing the ocean's resources, understanding its ecosystems, and ensuring sustainable maritime activities. However, the vast computational demands of analyzing complex marine data have historically impeded research progress. Nowadays, Artificial Intelligence (AI)-empowered oceanic computing is reshaping the solutions across various ocean-related research areas such as fluid dynamics, marine geology, ocean energy, autonomous underwater vehicle (AUV), and unmanned surface vehicle (USV). Specifically, AI's capabilities in modeling and predicting oceanic phenomena such as currents and waves are indispensable for precise weather forecasting and ensuring maritime safety. In marine geology, AI's ability to analyze ocean floor data unlocks new opportunities for identifying geological features and assessing mineral deposits, revolutionizing underwater exploration and resource extraction. The field of ocean energy is also witnessing a transformation as AI-enhanced methods improve the capture and conversion of oceanic forces into sustainable energy, promising alternative power solutions for coastal regions. The deployment of unmanned maritime vehicles, such as USVs and AUVs, equipped with AI for perception and navigation, is expanding the horizons of oceanographic research and environmental monitoring, making these operations safer and more efficient. Moreover, AI is integral to enhancing underwater robotics, enabling them to undertake various tasks, such as environmental conservation, archaeological investigations, and energy exploration. Building on the foundation laid by AI in revolutionizing oceanic computing, this special session aims to delve deeper into AI's transformative impact across pivotal marine domains. It seeks to highlight AI's integral role in advancing our comprehension, surveillance, preservation, and sustainable engagement with oceanic environments.

This session is dedicated to exploring state-of-the-art AI applications in oceanography and fostering cross-disciplinary collaborations among scientists, conservationists, and technologists. We will explore innovative solutions to overcoming the obstacles of integrating AI into marine research. Further, we aim to discuss AI's pivotal role in tackling critical issues like climate change effects, marine habitat preservation, and sustainable resource management. Through this session, we aim to highlight the ethical and sustainability questions related to AI's deployment in oceanographic endeavors.

Topics include (but are not restricted to):

  • Predictive Modeling for Oceanic Environmental
  • AI-Enhanced Monitoring for Marine Conservation
  • AI Innovations in Ocean Exploration and Surface Monitoring
  • Maritime Intelligent Maritime Transport System
  • Control Systems of Unmanned Maritime Vehicles
  • Multimodal Learning in Oceanic Computing
  • Maritime Situational Awareness System

5. Engineering applications of hybrid artificial intelligence tools

Led by Dr. Zbigniew Gomolka (University of Rzeszow), Prof. Ewa Dudek-Dyduch (AGH University of Krakow)

The interest in artificial intelligence leads to the consolidation of the activities of scientists and the education of the best experts in this field, so that the work of independent global centers can inspire their creators and find business applications faster. The development of simulation environments with a standardized API interface will allow for the collection of a large amount of data collected when interacting with the environment through the use of AI methods in the branches of management, automation, robotics, autonomous vehicles or energy consumption control. The use of fuzzy logic methods, evolutionary calculations and neural networks in intelligent decision support and control systems, including e.g. intelligent systems and machine learning methods for searching and processing information and supporting decision-making allow for optimal design of engineering systems. It seems important to use deep machine learning methods to recognize early symptoms of damage to physical objects based on the activity of their real processes, and to automatically detect anomalies in multidimensional production systems. Research on machine learning, statistical inference, and information theory, including variable selection methods in high-dimensional classification problems, will allow for smooth communication and detailed data exchange in algorithmic AI systems.

Topics include (but are not restricted to):

  • Learning strategy,
  • Distributed optimization algorithms design and analysis,
  • Data-based modeling and control for optimization complex system,
  • Intelligent technologies for optimizing discrete processes,
  • Intelligence technologies for Human–computer interaction,
  • Multi-task and multi-objective optimization,
  • Artificial intelligence applications for software engineering,
  • Knowledge management in software projects,
  • AI-centered Systems and Large-Scale Applications,
  • Evolutionary Algorithms and Evolutionary Computation,
  • Neural Networks and Deep Learning,
  • Hybrid and Hierarchical Intelligent Systems,
  • Hybrid artificial intelligence tools,
  • Multi-Agent Systems,
  • Knowledge Representation and Management,
  • Preprocessing of industry processes data for DNN,
  • AI for eyetracking technology,
  • Intelligent scheduling for discrete processes.

6. Computational Cognitive Neuroscience

Led by Dr. Zohreh Doborjeh (Auckland University of Technology), Prof. Alexander Sumich (Nottingham Trent University), Prof Paul Corballis (The University of Auckland), Assoc. Prof. Alan Wang (The University of Auckland), Prof Robin Palmer (Canterbury University) 

In the last decade, Artificial Intelligence (AI) has emerged as a transformative field, significantly advancing our ability to model and understand human cognitive and neural processes. By leveraging sophisticated algorithms, AI technologies can learn from complex datasets to extract meaningful patterns, facilitating tasks such as regression, prediction, and classification. These capabilities are particularly valuable in the fields of neuroscience and psychology, where understanding the intricate workings of the brain and mind requires analysing vast amounts of data from neuroimaging and behavioural studies.

This session aims to foster discussions on cutting-edge research and facilitate collaboration among researchers at the intersection of psychology, neuroscience, and computer science.

We invite submissions on the following topics:

  • Neuroimaging Studies: Applying advanced machine learning methods to brain imaging data (e.g., EEG, ERP, PET, fNIRS, sMRI, and fMRI) to understand and/or classify neurocognitive states.
  • Computational models of neural circuits: Modelling the neural mechanisms underlying sensory perception, cognitive functions, memory, learning, and decision making.
  • Mental Health and Neurological Disorders: Developing computational models for the prediction, intervention, and prevention of mental health and neurological disorders.
  • Theoretical Frameworks: Developing theoretical models to better understand cognitive phenomena.

7. Reliable, Robust and Secure Machine Learning Algorithms

Led by Dr Harry Nguyen (University College Cork - National University of Ireland, Cork), Dr Xuan-Son Vu, Monowar Bhuyan, Erik Elmroth (Umeå University), 

"The rise of machine learning (ML) and artificial intelligence (AI) makes many applications successful across societies, such as healthcare, finance, robotics, transportation and industry operations, by inducing intelligence in real-time. Designing, developing and deploying reliable, robust, and secure AI/ML algorithms are desirable for building trustworthy systems that offer trusted services to users with high-stakes decision-making. For instance, AI-assisted robotic surgery, automated financial trading, autonomous driving and many more modern applications are vulnerable to concept drifts, dataset shifts, misspecifications, misconfiguration of model parameters, perturbations, and adversarial attacks beyond human or even machine comprehension level, thereby posing dangerous threats to various stakeholders at different levels. Moreover, with the recent adoption of large language models (LLMs), AI/ML models are severely susceptible to reliable issues such as hallucinations and misinformation on a large scale. Therefore, building trustworthy AI systems requires lots of research efforts in addressing different mechanisms and approaches that could enhance user and public trust. To name a few, the following are known to be topics of interest in trustworthy and secure AI/ML/LLMs, but are not limited to: (i) bias and fairness, (ii) explainability, (iii) robust mitigation of adversarial attacks, (iv) improved privacy and security in model building, (v) being decent, (vi) model attribution and (vii) scalability of the model under adversarial settings. All of these topics are important and need to be addressed.

This special session aims to draw together state-of-the-art machine learning (ML) advances to address challenges for ensuring reliability, security and privacy in trustworthy systems. The challenges in different learning paradigms include but are not limited to: (i) robust learning, (ii) adversarial learning, (iii) stochastic, deterministic and non-deterministic learning, and (iv) secure and private large models. Nonetheless, all aspects of learning algorithms that can deal with reliable, robust and secure issues are the focus of the special session. It will focus on the robustness, performance guarantee, consistency, transparency and safety of AI, which is vital to ensure reliability. The special session will attract analytics experts from academics and industries to build trustworthy AI systems by developing and assessing theoretical and empirical methods, practical applications, and new ideas and identifying directions for future studies. Original contributions and comparative studies among different methods are welcome with an unbiased literature review.


 8. AI in Environmental, Conservation and Geospatial Applications

Led by Dr. Akbar Ghobakhlou, Prof. Jacqui Walley (Auckland University of Technology), Dr. Mike Watts, Dr. Asli Tece Bayrak (Media Design School)

Artificial intelligence (AI) offers promising solutions to critical environmental challenges, including climate change modelling, energy efficiency improvement, landscape erosion prediction, invasive species management, and endangered species conservation. In this context, computer vision techniques, particularly Convolutional Neural Networks (CNNs), play a pivotal role. These deep learning models enable us to extract meaningful information from geospatial data, satellite imagery, and environmental sensor networks. Geospatial data, often integrated with computer vision, provides insights for informed decision-making.

This special session is intended to attract papers dealing with the intersection between artificial intelligence and environmental issues.

Topics of interest include, but are not limited to:

  • Methods in the modelling of climate change and its effects
  • Predictive modelling of climate change impacts
  • Analyzing satellite imagery, LiDAR data, and other remote sensing sources
  • Habitat modeling and ecological niche analysis.
  • AI-driven species identification and tracking.
  • Conservation planning using machine learning algorithms
  • AI-driven approaches for biodiversity assessment
  • Remote sensing and geospatial analytics for environmental preservation
  • Precision farming techniques in crop yield prediction
  • Soil health assessment through AI techniques
  • Water resource management using machine learning
  • Geospatial data fusion for disaster management
  • Land cover classification, environmental monitoring

9. AI Education

Led by Dr. Michael Watts, Ranpreet Kaur (Media Design School), Dr. Akbar Ghobakhlou (Auckland University of Technology)

There has been an explosion in the applications of Artificial Intelligence (AI). While Large Language Models such as ChatGPT have garnered much of the attention, other AI technologies have also found wide application, such as the predictive keyboards on mobile devices, and facial recognition systems in supermarkets. Some technology venture capitalists have reported that 80% of the funding pitches they receive involve AI. Many business owners believe that AI is going to put them out of business, unless they adapt to the technology. Others are desperately searching for ways to get onto the AI bandwagon. This surge in interest in AI has led to a worldwide shortage of AI engineers. Furthermore, the inappropriate application of AI, whether through the use of biased data or unethical applications, has also led to social and economic fallout.
The increased public awareness of AI technologies has also led to a proliferation of media commentary, of varying degrees of competence, and governmental regulation. Some students have taken to using AI tools to assist in their assignments, while others have changed their career pathways due to a perception that AI is going to destroy their future job prospects.
There is, therefore, a need for education about AI. This need spans nearly all levels of education, from primary school through to postgraduates. At primary and secondary level so that people enter the working world with the basic knowledge of AI and how it affects their lives. At tertiary undergraduate and postgraduate level so that we have a steady supply of engineers and developers who can utilise AI in an appropriate and ethical manner.
This all raises a fundamental question: How is this education being done?
This special session is intended to attract papers dealing with all aspects of AI education. Topics of interest include, but are not limited to:

  • Incorporating AI into teaching curricula at all levels of education
  • The design and implementation of AI-specialist teaching curricula
  • Technologies used to teach AI
  • Teaching the ethics of AI
  • Policy making around AI education
  • The teaching of specialist topics within AI

10. Neural models of infants and child development

Led by Prof. Alistair Knott (Victoria University of Wellington), Prof. Annette Henderson (University of Auckland)

The dramatic advances of neural AI methods we have seen in the last few years are loosely based on the brain's distributed mode of computation, but are distinctively unhumanlike in the way they develop. LLMs, for instance, begin learning directly on vast quantities of unembodied mature adult language; it is only at a late stage that their learning is interactively shaped (by alignment) or becomes 'multimodal' (through interfaces with vision or action). By contrast, human infants' learning is fundamentally embodied: from birth, infants must learn to engage with the physical world, by meaningfully deploying their sensory and motor apparatus (Smith and Gasser, 2005). Infants' learning is also fundamentally staged, beginning with the acquisition of basic sensorimotor concepts and abilities, along with conceptions of close caregivers, and building on these (Vygotsky, 1994). Infants' learning is also interactive, driven by targeted real-time input from caregivers (Bornstein et al., 2008), but equally self-guided, driven by infants' own curiosity and experiences (Oudeyer et al., 2007).

There is a growing awareness that computational models of infant development may offer ways of augmenting the current generation of high-performing AI models. The session we propose will bring together researchers working on neural models of infant cognitive development, focussing on embodied learning, learning through interaction, self-guided learning, and staged learning. Crucially, the session will also invite participation from developmental psychologists. The work of psychologists studying development in human infants and children is newly relevant to work in AI, and their voices are increasingly heard in discussions about how AI should progress (see e.g. Smith, 2023; Gopnik and Chiang, 2024).

11. Computer Vision and Sensor Signal Processing for Enhancing Life Quality and Safety

Led by Assoc. Prof. Boris Bačić (Auckland University of Technology), Assoc. Prof. Nabin Sharma, Dr. Muhammad Saqib (University of Technology Sydney)

AI has made significant advancements in large language models, image, and video generation, while movement interpretation is still a challenge. In the post-pandemic years, global society has experienced an increase in cities density, with new challenges in balancing work-from-home and daily commutes, and other activities associated with an active lifestyle. This special session focuses on the neural information processing of movement data with applications of deep learning and computational intelligence to support society in finding ways to enhance the quality of life, an active lifestyle, and usability and safety of spaces where human activity occurs.

This special session aims to bring together academics and multidisciplinary experts collaborating in the field of computational sports science, sport analytics, rehabilitation and all experts interested in ways technology can support the health benefits of an active life.

The main topics of this special session include, but are not limited to the following:

  • Computer Vision (CV), pose estimation, and sensor signal processing for object detection, Human Activity Recognition (HAR), and Human Motion Modelling and Analysis (HMMA). Applications and technology for:
    • Explainable AI and hybrid approaches neural information processing
    • Video and sensor data fusion
    • Brain computer interfaces, EEG, EMG signal processing and modelling
    • Deep learning applications and architectures
    • Supervised, semi-supervised and unsupervised learning
    • Knowledge discovery and feature processing techniques
    • Open-source software modelling and development tools
  • Smart cities, spaces and infrastructures in support of:
    • Safety,
    • Utility,
    • Health, life quality, and active lifestyle
    • Advancements in wearable or sport-equipment attached technology
    • Augmented rehabilitation and coaching technology – enhancing rehabilitation, motor skills, human performance, and movement techniques
  • Advancements in motion assistive technologies (e.g. intelligent prosthetics rehabilitation and coaching devices). Case studies on:
    • Balance pattern modelling and analysis related to movement pathomechanics including fall prediction and detection
    • Exergames for regaining stability, control, motor skill, and technique adaptation
    • Technology-mediated challenges in post-surgery and brain or neuro-motor damage rehabilitation contexts (e.g. providing near-real time movement evaluation and diagnostics)
  • Human Computer Interaction (HCI) for augmented feedback and visualisation design e.g. via near-real time in 2D, 3D, portable computing devices or via Augmented/Immersive Reality (AR/IR)
  • Advancements in monocular, multi-camera and depth video processing for HAR, HMMA and object detection including integration with mobile, sensor signals, portable and low-cost consumer devices. 2D and 3D markerless pose estimation.
    Case studies involving:

    • Biomechanics parameters and raw data processing from video, wearables and mobile devices
    • Producing replay indexing and interaction e.g. via speech or gesture recognition, smartwatch, mobile or other devices
    • Enhancing media coverage: Game analytics, pattern detection from video and sensor signals, and visualisation. Data-driven game strategy analytics,
      information and visualisation
    • Motion pattern detection and indexing from video and sensor signals
    • Near real-time movement pattern detection and recognition from kinematic and kinetic data sources
    • Movement outcome prediction and cues extraction (e.g. tennis serve, shot and ball trajectory selection, cricket batting, or baseball pitching)
  • Privacy preservation filtering for:
    • safety monitoring
    • on-line exergaming,
    • rehabilitation
    • augmented coaching
    • health/elderly care. Case studies involving safety and privacy preservation, rehabilitation, exergames, sport or active life contexts including restricted, private or public environments where human motion activities occur.

Selected papers will be invited to upgrade their content to the special issue “Applications of Wearable Sensors in Healthcare, Rehabilitation and Sports Biomechanics” of the Applied
Sciences journal.