Computational Sciences and Artificial Intelligence in Industry

<ul> <li>Novel strategies for data-driven evolutionary optimization </li> <li>Machine learning using distance-based methods </li> <li>Counting cells and predicting immunoscore using gradient boosted convolutional neural networks </li> <li>Kubelka-Munk model and stochastic model comparison in skin physical parameter retrieval using neural networks </li> <li>A combined approach of neural networks and graphical models in skin cancer inference using spectral imaging </li> <li>Using wave propagation simulations and convolutional neural networks to retrieve thin coating’s thickness from hyperspectral images </li> <li>Predicting future overweight and obesity from childhood growth data: A case study </li> <li>Variable selection under a value acquisition budget </li> <li>Stochastic approximation by successive piecewise linearization </li> <li>Non-convex robust low-rank matrix recovery </li> <li>Neural network learning via successive piecewise linearization </li> <li>Learning for scientific computing purposes </li> <li>Computational intelligence in design of new nanomaterials </li> <li>Modeling flow, reactive transport and geomechanics in porous media </li> <li>Physics constrained machine learning for industrial applications </li> <li>Parameter and type identification in partial differential equations using deep neural networks </li> <li>Stability maximization for layered moving web with total mass constraint </li> <li>Similarity solutions for condensation on a non-isothermal vertical plate </li> <li>Enhanced topology optimization approach using moving morphable components coupled with NURBS curves</li> <li>Combined model order reduction and artificial neural network for data assimilation and damage detection in structures </li> <li>Towards the optimization of fuzzy pattern trees by abs - linearization </li> <li>Support vector machines in clusterwise linear regression</li> <li>A Second-order method with enriched hessian information for composite sparse optimization problems </li> <li>Missing value imputation via nonsmooth optimization and clusterwise linear regression</li> <li>Parsimonious neural networks </li> <li>Nobody can stop advancing artificial intelligence (AI) where developing </li> <li>Computational sciences, physics field theories and geometry </li> <li>Mini-symposium on ethics in AI </li> <li>Essentializing software engineering practices for ethically designing and developing artificial intelligence systems 30 Ethics is important, but how can we implement it? Survey on software developers’ views on AI ethics </li> <li>Industrial IoT capabilities in reducing the LCOE of offshore wind energy: A review</li> <li>High-Performance data analysis with the Helmholtz Analytics Toolkit (HeAT) </li> <li>&nbsp;Dynamic data-driven application systems based on tensor factorization: learning the physics of model evolution </li> <li>Predicting customer experience </li> <li>Puhti-AI: Finland’s new AI supercomputer </li> <li>Using Artificial Intelligence to Classify Textual Applications for Reporting Purposes&nbsp; Application of machine learning methods to error control of approximate solutions</li> <li>Iterative data selection strategy in offline data-driven evolutionary multiobjective optimization </li> <li>On surrogate management in interactive multiobjective building energy system design </li> <li>A modified deep neural network for the rapid inversion of geo-physical resistivity measurements </li> <li>Using agents for automatic meta-modelling algorithm selection in data-driven multiobjective optimization problems </li> <li>Future cooperation between Computational Science and AI in Industrial and Societal Applications - challenges, impact and expectations?</li> <li>Artificial Intelligence, Deep Learning and Science Policy in France </li> <li>AI and Digital Twin challenges in current EU arenas – in overall and from Finnish perspective</li> <li>AI and Data Analytics at a Centre for Scientific Computing </li> <li>AI in the field of medical applications </li> </ul> <p>&nbsp;</p>