Pricing Relational Data with Formal Guarantees

Data e Ora : Friday, March 24, 2017 - 11:00
Relatore : Prof. Paris Koutris
Affiliazione : University of Wisconsin-Madison, USA
Luogo : Sala riunioni DEI/G 318
Descrizione :

Abstract: Motivated by a growing market that involves buying and selling data over the web, we study pricing schemes that assign value to queries issued over a database. We present a formal framework for pricing queries over data that allows the construction of general families of pricing functions, with the main goal of avoiding arbitrage. Our main result is a complete characterization of the structure of pricing functions, by relating it to properties of a function over a lattice. We use our characterization, together with information-theoretic methods, to construct a variety of arbitrage-free pricing functions, and discuss various tradeoffs in the design space. Finally, we show how our framework can be implemented in practice, and perform query-based data pricing for a large class of SQL queries (including aggregations and join) in real time. This is joint work with Shaleen Deep.

Machine learning and complex networks for precision and systems biomedicine

Data e Ora : Thursday, March 2, 2017 - 10:30
Relatore : Prof. Carlo Vittorio Cannistraci
Affiliazione : Head of Biomedical Cybernetics Group, Technical University Dresden (Germany).
Luogo : Aula Magna "A. Lepschy"
Descrizione :

Abstract:The talk will present our research at the Biomedical Cybernetics Group that I established about three years ago in Dresden. We adopt a transdisciplinary approach integrating information theory, machine learning and network science to investigate the physics of adaptive processes that characterize complex interacting systems at different scales, from molecules to ecosystems, with a particular attention to biology and medicine. Our theoretical effort is to translate advanced mathematical paradigms typically adopted in theoretical physics (such as topology, network and manifold theory) to characterize many-body interactions in complex systems and quantitative biomedicine. We apply the theoretical frameworks we invent in the mission to develop computational tools for systems and network analysis. In particular, in biomedicine we deal with: prediction of wiring in biological networks, combinatorial and multiscale biomarkers design, precision biomedicine, drug repositioning and combinatorial drug therapy. In general, we devise theoretical models of structural organization in complex networks and we leverage this knowledge to create novel and more efficient algorithms and to perform advanced analyses and predictions of patterns in complex systems. This talk will focus on the Local Community Paradigm (LCP) which is a theory at the basis of epitopological learning in complex networks. LCP, which was inspired by Hebbian learning, was proposed to model local-topology-dependent link-growth in complex networks, therefore it is useful to devise topological methods for link prediction in monopartite and bipartite networks, but also as a topological measure to quantify the tendency of a network to be organized in local community. In particular, we will discuss the impact of LCP for pioneering topological methods for network-based drug-target interaction prediction and repositioning, and as a marker for the rewiring correlates of pain in the brain time-varying functional connectomes.  Biography:Carlo Vittorio Cannistraci is a theoretical engineer with a background in biomedical cybernetics. He is currently head of the Biomedical Cybernetics Group at the Centre for Molecular and Cellular Bioengineering (CMCB) and faculty member of the Department of Physics in the Technical University Dresden. His area of research embraces information theory, machine learning and complex network theory including also applications in computational network and systems biomedicine. Nature Biotechnology selected Carlo’s article (Cell 2010) on machine learning in developmental biology to be nominated in the list of 2010 notable breakthroughs in computational biology. Circulation Research featured Carlo’s work (Circulation Research 2012) on leveraging a cardiovascular systems biology strategy to predict future outcomes in heart attacks, commenting: “a space-aged evaluation using computational biology”. The Technical University Dresden honoured Carlo of the Young Investigator Award 2016 in Physics for his recent work on the local-community-paradigm theory and link prediction in bipartite networks.

Non-coherent massive MIMO

Data e Ora : Tuesday, January 31, 2017 - 11:00
Relatore : Dr Ana García Armada
Affiliazione : Universidad Carlos III de Madrid, Spagna
Luogo : DEI/D
Descrizione :

Massive MIMO has generated an enormous interest due to its many potential benefits in terms of capacity and energy efficiency. However, estimation and sharing of the large amount of channel state information may be a bottleneck for its deployment. In this talk, some ideas of non-coherent massive MIMO schemes will be discussed and a Non-Coherent Massive SIMO System based on M-DPSK and BICM-ID will be presented in the frame of the evolution of mobile communications towards 5G and beyond.

Distributed Computation of Large-Scale Graph Problems

Data e Ora : Thursday, January 12, 2017 - 15:30
Relatore : Michele Scquizzato
Affiliazione : Dept. of Computer Science, University of Houston TX USA
Luogo : Aula Magna T. Lepschy
Descrizione :

Abstract: Motivated by the need to understand the algorithmic foundations of distributed large-scale graph computations, we study some fundamental graph problems in a message-passing model for distributed computing. We present (almost) tight upper and lower bounds on the time complexity of several graph problems such as graph connectivity, minimum spanning tree, triangle enumeration, and PageRank. (Joint work with Gopal Pandurangan and Peter Robinson.)Biografia:Michele Scquizzato is a post-doctoral researcher in the Department of Computer Science at the University of Houston. His research focuses on the theory of parallel and distributed computing, algorithms for large-scale data analysis, and algorithmic problems related to energy-efficient computing. He received a Ph.D. in Information Engineering from the University of Padova in 2013