SUMO Lab focuses on the study and development of innovative software tools and techniques for efficient and accurate characterization, modeling, simulation and optimization of complex systems, with applications in different fields of science and engineering.
Our expertise has been successfully used by academic and business partners worldwide. We work closely with technology leaders on 'real world' challenges.
SUMO Lab performs basic, applied and contract research.
Surrogate models of complex systems
Surrogate models, or metamodels, are compact scalable analytic models that approximate the multivariate input/output behavior of complex systems, based on a limited set of computational expensive simulations. Surrogate models can be used for design automation, parametric studies, optimization and sensitivity analysis.
(Parametric) Macromodels of microelectronic and nanoelectronic systems
Macromodels are used to model and simulate the complex frequency behavior of passive electronic components and systems (RF and microwave components, antennas, broadband interconnects).
(Parameterized) Model Order Reduction (MOR/PMOR) techniques
Model Order Reduction (MOR) is a branch of system and control theory, which reduces the complexity of dynamical systems, while preserving (to the possible extent) their input-output behavior.
Supervised machine learning - Model Selection & Active Learning
Supervised learning is a Machine Learning (ML) technique for approximating the input/output behavior of complex systems. The task of the supervised learner is to predict the output behavior of a system for any set of input values, after an initial trainig phase.
In supervised learning, the selection of data samples (active learning) and the selection of models (model selection) is crucial for acquiring high accuracy surrogate models with good generalization capabilities.
Bioinformatics is the application of information technology to the field of molecular biology. Its primary goal is to increase our understanding of biological processes by developing and applying computationally intensive techniques (e.g., pattern recognition, data mining, machine learning algorithms, and visualization).
Currently, our research focuses on gene splicing, which is a very intricate and tightly regulated process in the cell. Different machine learning techniques (e.g. Feature Selection (FS), Instance Selection (IS), and mathematical models for gene splicing) are used to gain insight and to detect alternative splice variants.
High Performance Computing (HPC) - Scheduling algorithms for distributed computing
Application-aware scheduling algorithms can be used for efficiently distributing and managing parameterized computer experiments on multi-core processors, computer clusters and grids. The number of (computational expensive) simulations is automatically reduced.
Scientific computing is a rapidly growing multidisciplinary field that employs advanced computational techniques to understand and solve complex problems. Many problems in Computational Science and Engineering can be characterized by a "pipeline" that includes (mathematical) modeling techniques, simulation techniques (discretizations, algorithms, software frameworks, and problem solving environments), and analysis techniques (data mining, visualization, and error, sensitivity, stability, and uncertainty analyses).