Polytechnic University of Milan

About: Polytechnic University of Milan is a education organization based out in Milan, Italy. It is known for research contribution in the topics: Finite element method & Population. The organization has 18231 authors who have published 58416 publications receiving 1229711 citations. The organization is also known as: PoliMi & L-NESS.

Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks

Abstract: Matrix completion models are among the most common formulations of recommender systems. Recent works have showed a boost of performance of these techniques when introducing the pairwise relationships between users/items in the form of graphs, and imposing smoothness priors on these graphs. However, such techniques do not fully exploit the local stationary structures on user/item graphs, and the number of parameters to learn is linear w.r.t. the number of users and items. We propose a novel approach to overcome these limitations by using geometric deep learning on graphs. Our matrix completion architecture combines a novel multi-graph convolutional neural network that can learn meaningful statistical graph-structured patterns from users and items, and a recurrent neural network that applies a learnable diffusion on the score matrix. Our neural network system is computationally attractive as it requires a constant number of parameters independent of the matrix size. We apply our method on several standard datasets, showing that it outperforms state-of-the-art matrix completion techniques.

Product lifecycle management – from its history to its new role

Abstract: This paper is a result of comprehensive consultation among the authors, with the scientists and leading actors in the area of PLM, which is a reference term for a list of phenomena currently ongoing in the industrial community. This paper discusses the pervasive concept of product lifecycle management (PLM), starting from its history to its constituent elements and its role in the current industry. The authors propose and elaborate their vision for the future steps of the PLM in terms of emerging issues and topics that industrial practitioners and researchers need to address.

Generalized molecular orbital tomography

Abstract: Atomic and molecular gases generate extreme ultraviolet light when excited by pulses of intense laser light. This emission provides information about the inner workings of the molecules and even enables us to map electron orbitals. However, so far molecular orbital tomography has been restricted to simple molecules. A technique that can be applied to more complicated molecules is now unveiled.

Satellite-based measurements of surface deformation reveal fluid flow associated with the geological storage of carbon dioxide

Abstract: Interferometric Synthetic Aperture Radar (InSAR), gathered over the In Salah CO{sub 2} storage project in Algeria, provides an early indication that satellite-based geodetic methods can be effective in monitoring the geological storage of carbon dioxide. An injected volume of 3 million tons of carbon dioxide, from one of the first large-scale carbon sequestration efforts, produces a measurable surface displacement of approximately 5 mm/year. Using geophysical inverse techniques we are able to infer flow within the reservoir layer and within a seismically detected fracture/ fault zone intersecting the reservoir. We find that, if we use the best available elastic Earth model, the fluid flow need only occur in the vicinity of the reservoir layer. However, flow associated with the injection of the carbon dioxide does appear to extend several kilometers laterally within the reservoir, following the fracture/fault zone.