<p><strong>Abstract.</strong> The use of mobile computing technologies can change the experience of visiting cultural sites by making vast digital heritage collections accessible on site. The spread of machine learning technologies on mobile devices is encouraging the interaction of artificial intelligence with the shape of the built environment. However, while some research already applies deep learning image recognition in an urban context, the literature on how to develop effective neural networks to detect architectural features is still limited, as well as the availability of architecture-related datasets. This work presents the steps and results of the prototype development of a mobile app to perform monument recognition using convolutional neural networks. The tool allows users to interact with the physical space and access a digital archive of texts, models, images and other data.</p>

We study the numerical simulation of the shaken dynamics, a parallel Markovian dynamics for spin systems with local interaction and transition probabilities depending on the two parameters q and J that “tune” the geometry of the underlying lattice. The analysis of the mixing time of the Markov chain and the evaluation of the spin-spin correlations as functions of q and J, make it possible to determine in the (q, J) plane a phase transition curve separating the disordered phase from the ordered one. The relation between the equilibrium measure of the shaken dynamics and the Gibbs measure for the Ising model is also investigated. Finally two different coding approaches are considered for the implementation of the dynamics: a multicore CPU approach, coded in Julia, and a GPU approach coded with CUDA.

We perform a numerical investigation of the shaken dynamics, a parallel Markovian dynamics for spin systems with local interaction and whose transition probabilities depend on two parameters, q and J, that tune the geometry of the underlying lattice. We determine a phase transition curve, in the (q, J) plane, separating the disordered phase from the ordered one, study the mixing time of the Markov chain and evaluate the spin-spin correlations as q and J vary. Further, we investigate the relation between the equilibrium measure of the shaken dynamics and the Gibbs measure for the Ising model. Two different approaches are considered for the implementation of the dynamics: a multicore CPU approach, with code written in Julia and a GPU approach with code written in CUDA.