The algorithmic filtering of contents exchanged on digital social networks entails a centralized control of the intensity of users' interaction. The aim of this article is to investigate how this centralized action can affect the time evolution of users' opinions on a specific issue. To this purpose, this article proposes a stochastic multiagent model that incorporates a simplified description of the modulation of the interaction intensity exerted by the platform manager. Various transient and steady-state properties of the model are established. In particular, it is studied how emerging collective behaviors, e.g., consensus, polarization, and community cleavage, depend on the interaction intensity parameter. Notably, the nonmonotonic effects of such a parameter are observed for suitable distributions of influenceability across the community. By means of a newly introduced concept of individual stochastic social power, some insights are given on the role of the interaction intensity parameter in the conflict between opposite factions in simple scenarios. A major finding of this article is that an apparently neutral intervention, i.e., unbiased with respect to the conflicting opinions, can favor one faction just by tuning the interaction intensity.

In this paper, the calibration of the parameters of the Gordon-Ng Universal (GNU) chiller model is investigated. In its standard formulation, the GNU model is written as a linear-in-parameter model that can be calibrated by Ordinary Least Squares. It has been already observed elsewhere that, since the regressors are subject to measurement inaccuracies, the OLS approach is prone to yield biased estimates of the parameters. As a remedy, Andersen and Reddy proposed the adoption of an Errors in Variable (EIV) framework, showing that bias could be reduced or even eliminated by means of a corrected least squares algorithm. However, some questions remained open. Given that the EIV approach achieves bias reduction at the cost of increasing the variance, is it really preferable to OLS? If the final goal is not parameter estimation, but the prediction of the Coefficient of Performance (COP), how does OLS compare with EIV? And what is the most appropriate calibration method, under a statistical viewpoint? Finally, is the added complexity of a statistically rigorous approach employing Nonlinear Least Squares (NLS) really worth the potential improvements in COP prediction? In order to answer these questions, three estimation methods, OLS, EIV and NLS, are tested on two benchmarks: a public precise chiller performance dataset and an ASHRAE dataset. The results suggest, that OLS estimation, in spite of its suboptimality, may prove largely satisfactory both for parameter estimation and COP prediction, although it may be worth analyzing other more challenging COP prediction problem before the final word is said.

A client-server architecture to simultaneously solve multiple learning tasks from distributed datasets is described. In such architecture, each client corresponds to an individual learning task and the associated dataset of examples. The goal of the architecture is to perform information fusion from multiple datasets while preserving privacy of individual data. The role of the server is to collect data in real time from the clients and codify the information in a common database. Such information can be used by all the clients to solve their individual learning task, so that each client can exploit the information content of all the datasets without actually having access to private data of others. The proposed algorithmic framework, based on regularization and kernel methods, uses a suitable class of "mixed effect" kernels. The methodology is illustrated through a simulated recommendation system, as well as an experiment involving pharmacological data coming from a multicentric clinical trial.

Standard single-task kernel methods have recently been extended to the case of multitask learning in the context of regularization theory. There are experimental results, especially in biomedicine, showing the benefit of the multitask approach compared to the single-task one. However, a possible drawback is computational complexity. For instance, when regularization networks are used, complexity scales as the cube of the overall number of training data, which may be large when several tasks are involved. The aim of this paper is to derive an efficient computational scheme for an important class of multitask kernels. More precisely, a quadratic loss is assumed and each task consists of the sum of a common term and a task-specific one. Within a Bayesian setting, a recursive online algorithm is obtained, which updates both estimates and confidence intervals as new data become available. The algorithm is tested on two simulated problems and a real data set relative to xenobiotics administration in human patients.

The commonality analysis is a proven tool for fault detection in semiconductor manufacturing. This methodology extracts subsets of production lots from all the available data. Then, data mining techniques are used only on the selected data. This approach loses part of the available information and does not discriminate among the lots. The new methodology performance the automatic classification of the electrical wafer test maps in order to identify the classes of failure present in the production lots. Subsequently, the proposed procedure uses the process history of each wafer to create a list of the root cause candidates. This methodology is the core of the software tool ACID which is currently used for process diagnosis at the Agrate site of the ST Microelectronics. A real analysis is presented.

Regularization networks are nonparametric estimators obtained from the application of Tychonov regularization or Bayes estimation to the hypersurface reconstruction problem. Their main drawback back is that the computation of the weights scales as O(n(3)) where n is the number of data. In this paper, we show that for a class of monodimensional problems, the complexity can be reduced to O(n) by a suitable algorithm based on spectral factorization and Kalman filtering. Moreover, the procedure applies also to smoothing splines.

The paper deals with the problem of reconstructing a continuous 1D function from discrete noisy samples. The measurements may also be indirect in the sense that the samples may be the output of a linear operator applied to the function. Bayesian estimation provides a unified treatment of this class of problems. We show that a rigorous Bayesian solution can be efficiently implemented by resorting to a Markov chain Monte Carlo (MCMC) simulation scheme. In particular, we discuss how the structure of the problem can be exploited in order to improve the computational and convergence performances. The effectiveness of the proposed scheme is demonstrated on two classical benchmark problems as well as on the analysis of IVGTT (IntraVenous glucose tolerance test) data, a complex identification-deconvolution problem concerning the estimation of the insulin secretion rate following the administration of an intravenous glucose injection.

In this article we deal with the problem of interpreting data coming from a dynamic system by using causal probabilistic (CPN), a probabilistic graphical model particularly appealing in Intelligent Data Analysis. We discuss the different approaches presented in the literature, outlining their pros and cons through a simple training example. Then, we present a new method for reconstructing the state of the dynamic system, based on Markov Chain Monte Carlo algorithms, called dynamic probabilistic network smoothing (DPN-smoothing). Finally, we present an example of the application of DPN-smoothing in the field of signal deconvolution.

(See IEEE Trans. Signal Processing, vol.40, p.2377-2391, 1992) surveyed several control methods for rejection of narrow-band disturbances. In particular, the authors introduced a method, called disturbance modeling (DM), which is a linear quadratic (LQ) optimization technique hinging on a state-space model of the disturbance. In the present paper, it is demonstrated that the DM design methodology proposed by Sievers and von Flotow leads to an unsolvable optimization problem. Furthermore, it is suggested how to modify the LQ cost functional in order to obtain an effective DM disturbance rejection scheme. Duality between DM and the approach based on frequency shaped cost functionals is concisely discussed.>