You are here: Home / Research / Main research areas / Computer sciences

Computer sciences

The Department of Mathematics and Computer Science has a strong commitment for research and teaching in the field of Computer Science and related areas.

The research activity focuses on the following broad subjects:

  • design and optimization of high performance scientific codes for massively parallel and distributed computing systems;
  • evaluation, benchmarking, modelling and profiling of performances of HPC architectures;
  • simulation at large scale of complex system: spin glass using Monte Carlo algorithms, fluido-dynamics using lattice Boltzmann methods, and lattice quantum chromodynamics;
  • development and evolution of distributed computing infrastructures like grids and clouds;
  • advanced information systems: relational and noSQL databases, distributed and large scale data storage, time-series analytics;
  • design and development of data-acquisition systems for measurements and instrument control and real-time processing of scientific images;
  • development of scientific software, targeted to both scalar and HPC platforms, implementing numerical methods for large-scale, nonlinear, possibly non-smooth optimization problems. Applications include: discrete optimal control, statistical learning techniques and inverse problems;
  • analysis of filtering techniques for big data and applications to bioinformatics and medical imaging in HPC framework;
  • knowledge representation and reasoning: probabilistic logic programming, knowledge compilation, Monte Carlo approximate inference, lifted inference;
  • uncertainty in the semantic web: probabilistic description logics, inference and learning;
  • data mining and machine learning combining relational and statistical aspects;
  • big data and techniques for supporting large scale data analytics: massively distributed and parallel algorithms on clusters and clouds;
  • linked open data and automated retrieval of relevant knowledge for machine learning;
  • non-classical logics: decidability, complexity and inference procedures for modal, temporal and description logics.