Thematic tracks organized by experts in a particular area constitute the core of the conference.
The list of accepted tracks is below, please click through for brief information and track web/contact addresses to follow to find full details.
We will be adding several more tracks in the coming weeks.
If you are interested in organizing a thematic track at ICCS 2021, you can find all necessary details on the Call for Tracks webpage.
- Advances in High-Performance Computational Earth Sciences: Applications and Frameworks – IHPCES
- Applications of Computational Methods in Artificial Intelligence and Machine Learning – ACMAIML
- Architecture, Languages, Compilation and Hardware support for EMerging and Heterogeneous sYstems – ALCHEMY
- Artificial Intelligence and High-Performance Computing for Advanced Simulations – AIHPC4AS
- Biomedical and Bioinformatics Challenges for Computer Science – BBC
- Classifier Learning from Difficult Data – CLD2
- Computational Analysis of Complex Social Systems – CSOC
- Computational Collective Intelligence – CCI
- Computational Health – CompHealth
- Computational Methods for Emerging Problems in (dis-)Information Analysis – DisA
- Computational Methods in Smart Agriculture – CMSA
- Computational Optimization, Modelling and Simulation – COMS
- Computational Science in IoT and Smart Systems – IoTSS
- Computer Graphics, Image Processing and Artificial Intelligence – CGIPAI
- Data-Driven Computational Sciences – DDCS
- Machine Learning and Data Assimilation for Dynamical Systems – MLDADS
- MeshFree Methods and Radial Basis Functions in Computational Sciences – MESHFREE
- Multiscale Modelling and Simulation – MMS
- Quantum Computing Workshop – QCW
- Simulations of Flow and Transport: Modeling, Algorithms and Computation – SOFTMAC
- Smart Systems: Bringing Together Computer Vision, Sensor Networks and Machine Learning – SmartSys
- Software Engineering for Computational Science – SE4Science
- Solving Problems with Uncertainty – SPU
- Teaching Computational Science – WTCS
- Uncertainty Quantification for Computational Models – UNEQUIvOCAL
Advances in High-Performance Computational Earth Sciences: Applications and Frameworks – IHPCES
Topics of interest include, but not limited to:
- Large-scale simulations on both homogeneous and heterogeneous supercomputing systems in earth sciences, such as atmospheric science, ocean science, solid earth science, and space & planetary science, as well as multi-physics simulations.
- Advanced modeling and simulations on natural disaster prevention and mitigation.
- Advanced numerical methods such as FEM, FDM, FVM, BEM/BIEM, Mesh-Free method, and Particle method etc.
- Parallel and distributed algorithms and programming strategies focused on issues such as performance, scalability, portability, data locality, power efficiency and reliability.
- Software engineering and code optimizations for parallel systems with multi-core processors, GPU accelerators or Xeon Phi processors.
- Algorithms for Big Data analytics and applications for large-scale data processing such as mesh generation, I/O, workflow, visualization and end-to-end approaches.
- Methodologies and tools designed for extreme-scale computing with emphasis on integration, interoperability and hardware-software co-design.
Applications of Computational Methods in Artificial Intelligence and Machine Learning – ACMAIML
A major difficulty in dealing with modern data is that many of the old methods that have been developed for analyzing data during the last decades cannot be applied directly to modern data. One major solution, to overcome this challenge, is to effectively use ensemble methods such as deep artificial neural networks. These types of ensemble models are heavily reliant on the deployment of efficient computational methods. Thus, it’s even more imperative to deploy faster, more accurate and robust computational techniques for AI and ML models.
This track covers the application of computational methods for Artificial Intelligence and Machine Learning models.
Architecture, Languages, Compilation and Hardware support for EMerging and Heterogeneous sYstems – ALCHEMY
Artificial Intelligence and High-Performance Computing for Advanced Simulations – AIHPC4AS
Biomedical and Bioinformatics Challenges for Computer Science – BBC
The aim of this workshop is to bring together computer science and life scientists to discuss emerging and future directions in topics related to key bioinformatics and computational biology techniques: (i) Advanced computing architectures; (ii) Algorithm design; (iii) Data analysis and knowledge discovery; (iv) Data management and integration; (v) Integration of quantitative/symbolic knowledge into executable biomedical “theories” or models.
A special session will be devoted to bioinformatics and computer science methods to fight the COVID-19 pandemics.
Classifier Learning from Difficult Data – CLD2
The main aim of this workshop is to discuss the problems of data difficulties, to identify new issues, and to shape future directions for research.
Topics include (but not limited to):
- learning from imbalanced data
- learning from data streams, including concept drift management
- learning with limited ground truth access
- learning from high dimensional data
- learning with a high number of classes
- learning from massive data, including instance and prototype selection
- learning on the basis of limited data sets, including one-shot learning
- learning from incomplete data
- case studies and real-world applications
Computational Analysis of Complex Social Systems – CSOC
The aim of the workshop is to stimulate interdisciplinary research and cooperation to develop complex systems-based approaches aimed at understanding social systems. In general, the workshop will focus on the following questions: What are the patterns in social systems that cannot be explained by the existing theories and data? What kinds of data are needed to better inform the models? What new modeling techniques and methods need to be developed?
One of the immediate outputs of the workshop will be the publication of a special issue in a scientific journal to be determined.
Broad topics in social complexity:
- Modeling social phenomena such as (but not limited to) – population dynamics, migration patterns, armed conflicts, political movements, natural disasters, etc – and the many possible arrangements of relationships between these discrete phenomena.
- Modeling for sustainable development goals (poverty, well-being, food security, water, energy, etc.) in rapidly growing urban complex societies. Focusing on the role of modeling in policy and urban planning.
- Modeling individual and group behavior. New modeling techniques in computational social science.
- Emergent phenomena of complex social systems: self-organization, regime shifts, tipping points, and resilience.
- Empirical calibration and validation of computational models for complex social systems.
- Novel (curated) datasets for understanding complex social systems. The Social Complexity of Digital Data (new metatheory, complex realism in social research).
- Computational analysis of complex social systems, e.g., network analysis, social media, big data, etc.
Computational Collective Intelligence – CCI
The session is organized by the IEEE Computational Collective Intelligence technical committee, but welcomes all interested parties. The goal is to provide an internationally respected forum for scientific research in the computer-based methods of collective intelligence and their applications in (but not limited to) such fields as semantic web, social networks and multi-agent systems.
The topics of interest include, but are not limited to:
- Group decision making
- Consensus computing
- Collective action coordination
- Inconsistent knowledge processing
- Ontology mapping and merging
- Collaborative ontology
- Ontology development in social networks
- Semantic social networks
- Semantic and knowledge grids
- Semantic annotation of web data resources
- Group web services (service description, discovery, composition)
- Automatic metadata generation
- Semantic web inference schemes
- Reasoning in the semantic web
- Knowledge portals
- Knowledge discovery
- Information retrieval
- Advanced analysis for social networks dynamics
- Social networks and semantic communication
- Cooperative distributed problem solving
- Multiagent planning
- Negotiation protocols
- Multiagent learning
Computational Health – CompHealth
The scope of the workshop includes (but not limited to) the following areas: simulation and modeling in H&M; complex processes and systems in H&M; networks in H&M; uncertainty management in H&M; numerical methods in H&M; data & process mining, ML & AI in H&M; knowledge and data processing in H&M; decision support and recommending systems in H&M; advanced medical information systems; etc.
Computational Methods for Emerging Problems in (dis-)Information Analysis – DisA
However information spreading can be also used for disinformation. The problem of the fake news publication is not new and it already has been reported in ancient ages, but it has started having a huge impact especially on social media users or people watching media news (Internet, newspapers, tv etc.). Such false information should be detected as soon as possible to avoid its negative influence on the readers and in some cases on their decisions.
Another problem and emerging challenge is coming from using automated information analysis and decision support systems (based on Artificial Intelligence (AI) and Machine Learning (ML) advances) as black-box single truth providers. If those information analysis systems are misused, attacked or fooled, their results will also lead to (dis-) information.
The main aim of this workshop is to bring together researchers and scientists computational science who are pioneering (dis-)information analysis methods to discuss problems and solutions in this area, to identify new issues, and to shape future directions for research. Moreover, we invite prospective researchers to send papers concerning (dis-)information detection methods and architectures, explainability of information processing methods and decision support systems as well as their security.
The list of possible topics includes, but is not limited to:
- computational methods for (dis-) information analysis, especially in heterogenous types of data (images, text, tweets etc.)
- detection of fake news detection in social media
- deepfakes analysis
- images and video manipulation recognition
- architectural frameworks and design for (dis-)information detection
- aspects of explainability of information analysis systems and methods (including explainability of ML)
- adversarial attacks on information analysis
- explainability of deep learning
- learning how to detect the fake news in the presence of concept drift
- learning how to detect the fake news with limited ground truth access and on the basis of limited data sets, including one-shot learning
- proposing how to compare and benchmark the fake news detectors
- case studies and real-world applications
- human rights, legal and societal aspects of (dis-)information detection, including data protection and GDPR in practice
The session will be supported by by SocialTruth project (Socialtruth.eu), which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 825477 and will be technically endorsed by IEEE SMC TC on Big Data Computing.
Computational Methods in Smart Agriculture – CMSA
Topics include, but are not limited to:
- Optimisation in agro-ecosystems
- Intelligent irrigation systems
- Integrated sensing technology
- Precision agriculture methods
- Agriculture decision support systems
The area presents significant and challenging difficulties in developing robust predictive tools subject to uncertainties in climate and ecosystem responses, with recommendations needed that are relevant, productive and accurate over multiple decades. Furthermore, expertise in agronomy and economic factors needs to be integrated with modelling and simulation to allow analysis of “whole system” dynamics. Current research efforts are proceeding in algorithm development and innovative computational methods to solve these problems, and deliver outcomes. In the context of an interconnected world – the research has the potential to contribute to climate change adaptation and food security in a global context. Research and tools are needed for decision support for policy responses at a range of levels of responsibility: individual, regional, national and international.
Computational Optimization, Modelling and Simulation – COMS
COMS2021 will focus on new algorithms and methods, new trends, and latest developments in computational optimization, modelling and simulation as well as applications in science, engineering and industry.
Topics include (but not limited to):
- Computational optimization, engineering optimization and design
- Bio-inspired computing and algorithms
- Metaheuristics (ant and bee algorithms, cuckoo search, firefly algorithm, genetic algorithms, PSO, simulated annealing etc)
- Simulation-driven design and optimization of computationally expensive objectives
- Surrogate- and knowledge-based optimization algorithms
- Scheduling and network optimization
- Integrated approach to optimization and simulation
- Multiobjective optimization
- New optimization algorithms, modelling techniques related to optimization
- Design of experiments
- Application case studies in engineering and industry
Computational Science in IoT and Smart Systems – IoTSS
Computer Graphics, Image Processing and Artificial Intelligence – CGIPAI
Potential topics include, but are not limited to:
- Geometric and Solid Modeling & Processing, CAD/CAM/CAE, Curve/Surface Reconstruction
- Computer Graphics Techniques, Algorithms, Software and Hardware
- Computer Animation and Video Games
- Virtual and Augmented Reality, Virtual Environments and Autonomous Agents
- Computer Graphics Applications (Science, Engineering, Education, Health, Industry, Entertainment)
- Image Processing techniques
- Image Processing processes (e.g., image denoising, image deblurring, image segmentation, image reconstruction, depth estimation, 3D surface restoration)
- Image Processing applications
- Evolutionary and Nature-Inspired Algorithms (evolutionary programming, genetic algorithms)
- Neural Networks, Machine Learning, Deep Learning and Data Mining
- Swarm Intelligence and Swarm Robotics
- Bio-informatics and bio-engineering
- Natural Computing, Soft Computing and Evolutionary Computing
- Artificial Intelligence Applications
- Interplay among some of the previous areas.
Data-Driven Computational Sciences – DDCS
A data-driven computational system is the integration of a simulation with dynamically and intelligently assimilated data, multiscale modeling, computation, and a two way interaction between the model execution and the data acquisition methods (see the DDDAS Scientific Community Web Site, http://www.dddas.org). The workshop will present opportunities as well as challenges and approaches in technology needed to enable Data-Driven Computational Science capabilities in applications, relevant algorithms, and software systems. All related areas in Data-Driven Sciences are included in this workshop, including CyberPhysical Systems like HealthKit on iPhones and iPads as well as similar systems developed by Intel, Google, and Microsoft for phones and tablets, Internet of Things (IoT), Cloud of Things (CoT), and Data Intensive Scientific Discovery (DISD).
A recent example is a tranformative way of landing airplanes on time and reduce delays and cancellations is a process known as Time Based Flow Systems (TBFS) [UKNATS]. It spaces planes by space instead of by time. The first of these systems was developed for Heathrow Airport by Lockheed Martin for the British National Air Traffic Services and fully deployed in May, 2015. It has reduced flight cancellations due to wind by exactly 100% and flight delays by approximately 40% during the period of May – August, 2015.
Machine Learning and Data Assimilation for Dynamical Systems – MLDADS
The intersection of the fields data assimilation, machine learning and dynamical systems is largely unexplored, and the goal of the MLDADS workshop is to bring together contributions from these fields to fill the gap between these theories in the following directions:
(1) Machine Learning for Data Assimilation: how to assist or replace the traditional methods in making forecasts, without the unrealistic assumption (particularly linearity, normality and zero error covariance) of the conventional methods.
(2) Machine Learning for Dynamical Systems: how to analyze dynamical systems on the basis of observed data rather than attempt to study them analytically.
(3) Data Assimilation for Machine Learning and/or Dynamical Systems: how well does the model under consideration (Machine Learning model and/or Dynamical System) represent the physical phenomena.(4) Data Assimilation and Machine Learning for Dynamical Systems: how can tools from the interaction between the theories of Data Assimilation and Machine Learning be used to improve the accuracy of the prediction of dynamical systems.
MeshFree Methods and Radial Basis Functions in Computational Sciences – MESHFREE
The meshfree methods are especially convenient for scattered data processing as they do not require tessellation. They are used not only for interpolation and approximation, but also for a solution of partial and ordinary differential equations, etc. Meshfree methods are scalable to higher dimensions and offer smooth final representation and they lead to a solution of a system of linear equations, in general.
This ICCS 2021 workshop is intended to explore broad computational applicability of the Meshfree methods especially based on Radial Basis Functions across many areas, including theoretical and mathematical aspects of the Meshfree methods.
The aim is also to connect latest theoretical research results with possible computational applications, i.e. put together theory and applications in computational sciences.
Main topics (but not limited to):
- Meshfree methods in engineering problems
- Meshfree methods and differential equations
- Meshfree methods and GIS, CAD/CAM systems
- Meshfree methods in theory and practice
- Meshfree methods and computational and numerical issues
- Meshfree interpolation and approximation methods for large scalar and vector data sets
- Meshfree methods for scattered spatio-temporal data, t-varying systems etc.
- Radial Basis Functions (RBF) in computer graphics, visualization etc.
- Meshfree methods in image processing and computer vision
- Meshfree methods and projective space representation
- Comparison of meshfree and mesh based computational methods
- Scattered data interpolation and approximation methods
- Radial Basis Functions for a mesh morphing and data mapping
- Meshfŕee methods for corrupted image reconstruction and inpainting removal
- Meshfree methods applications in general
Multiscale Modelling and Simulation – MMS
The multiscale modelling and simulation (MMS) workshop aims to provide a forum for multiscale application modellers, framework developers and experts from the distributed infrastructure communities to identify and discuss challenges in, and possible solutions for, modelling and simulating multiscale systems, as well as their execution on advanced computational resources and their validation against experimental data.
Key topics of the MMS workshop include:
- Simulation and modelling of multiscale systems.
- Challenging applications in science, industry or society (e.g. in computational biology).
- Verification, validation and uncertainty quantification in a multiscale simulation/modelling context.
- New approaches for coupling and scale bridging, to combine different models and scales in one application.
- Advanced numerical methods for solving multiscale problems.
- Software approaches for simulating multiscale systems, and handling the complex workloads accompanying it.
- Executing multiscale models on advanced computational infrastructures (distributed, HPC, cloud, etc.).
- Performance analysis of multiscale applications and/or tools.
Quantum Computing Workshop – QCW
Simulations of Flow and Transport: Modeling, Algorithms and Computation – SOFTMAC
The international workshop on “Simulations of Flow and Transport: Modeling, Algorithms and Computation” (SOFTMAC) has been held 9 years since 2011 within the International Conference on Computational Science (ICCS). The aim of this special issue is to bring together researchers in the aforementioned field to highlight the current developments both in theory and methods, to exchange the latest research ideas, and to promote further collaborations in the community. We invite original research articles as well as review articles describing the recent advances in mathematical modeling, computer simulation, numerical analysis, and other computational aspects of flow and transport phenomena of flow and transport. Potential topics include, but are not limited to:
- advanced numerical methods for the simulation of subsurface and surface flow and transport, and associated aspects such as discretization, gridding, upscaling, multiscale algorithms, optimization, data assimilation, uncertainty assessment, and high performance parallel and grid computing;
- spatial discretization schemes based on advanced finite element, finite volume, and finite different methods; schemes that preserve local mass conservation (such as mixed finite element methods and discontinuous Galerkin methods) are of particular interest;
- decomposition methods for improved efficiency and accuracy in treating flow and transport problems; decomposition methods for nonlinear differential equations and dynamical systems arising in flow and transport; temporal discretization schemes for flow and transport;
- a-priori and a-posteriori error estimates in discretizations and decompositions; numerical convergence study; adaptive algorithms and implementation;
- modeling and simulation of single-phase and multi-phase flow in porous media or in free space, and its applications to earth sciences and engineering;
- modeling and simulation of subsurface and surface transport and geochemistry, and its application to environmental sciences and engineering;
- computational thermodynamics of fluids, especially hydrocarbon and other oil reservoir fluids, and its interaction with flow and transport;
- computational modeling of flow and transport in other fields, such as geological flow/transport in crust and mantle, material flow in supply chain networks, separation processes in chemical engineering, information flow, biotransport, and intracellular protein trafficking, will also be considered.
Smart Systems: Bringing Together Computer Vision, Sensor Networks and Machine Learning – SmartSys
Software Engineering for Computational Science – SE4Science
Specifically, we are interested in the software development and software engineering challenges and enablers relating to: (1) Computational science software applications solve complex software- or data-intensive research problems. These applications range from large parallel models/simulations of the physical world using HPC systems to smaller scale simulations developed by a single scientist or engineer on a desktop machine or a small cluster. (2) Applications domains ranging from humanities to engineering to science. (3) Applications that support scientific research and experiments at scale. Such applications include, but are not limited to, systems for managing and/or manipulating large amounts of data and systems that provide infrastructure for scientific or engineering applications such as libraries or HPC/Cloud software. (4) The process for building, reusing, and publishing software and data used in scientific experiments or engineering innovations. Among others, these processes include agile approaches, open source/open data issues, testing scientific software, and managing software or data repositories for publishing goals.
This workshop will build upon previous SE4Science workshops. Similar to the format of the previous workshops, in addition to presentation and discussion of the accepted papers, we plan to devote significant time during the workshop to discussing important topics that arise from the paper presentations. The goal of these discussions is to (1) develop a joint research plan that can be conducted collectively by workshop participants and (2) development of ideas/draft of position statements to be published externally.
Solving Problems with Uncertainty – SPU
With the advent of exascale computing, big data analytics and scalable AI, larger and larger problems – often requiring hybrid approaches – have to be tackled in a systematic way at scale. The variety of methods and approaches of solving such problems with uncertainties and quantifying the uncertainties become even more important due to the variety and scale of uncertainties in such problems.
The focus of the workshop will be on methods and algorithms for solving problems with uncertainties, stochastic methods and algorithms for solving problems with uncertainties, methods and algorithms for quantifying uncertainties such as dealing with data input and missing data, sensitivity analysis (local and global), dealing with model inadequacy, model validation and averaging, software fault-tolerance and resilience and more.
Teaching Computational Science – WTCS
Uncertainty Quantification for Computational Models – UNEQUIvOCAL
- Forward and inverse uncertainty quantification;
- Model inadequancy;
- Sensitivity analysis;
- Dimension reduction;
- Surrogate modelling (including machine-learning techniques and reduced-order modelling);
- Case studies showing efficient uncertainty quantification methods;
- Software for uncertainty quantification.