ANGUITA , Davide

This talk outIines sorne ofthe ideas and discussions carried on by RTD-SAS researchers during the first year of the EUNITE network. The focus is mainly on theoretical aspects of Smart Adaptive Systems that will serve as the basis for the creation of successful applications. Smart Adaptive Systems are of paramount importance in many application...

Algorithms, applications and hardware implementations of neural networks are not investigated in close connection. Researchers working in the development of dedicated hardware implementations de- velop simplied versions of otherwise complex neural algorithms or de- velop dedicated algorithms: usually these algorithms have not been thor- oughly...

We examine the efficient implementation of back-propagation (BP) type algorithms on T0, a vector processor with a fixed-point engine, designed for neural network simulation. Using Matrix Back Propagation (MBP) we achieve an asymptotically optimal performance on T0 (about 0.8 GOPS) for both forward and backward phases, which is not possible with...

We apply here a probabilistic method to predict the effect of quantizing the parameters of a Support Vector Machine. Thank to the particular structure of the SVM, the dependency of the output from the quantization noise can be predicted with good accuracy, and a simple closed–form formula can be derived, without imposing any hard–to–verify assumption

In the last few years several kinds of recurrent neural networks (RNNs) have been proposed for solving linear and nonlinear optimization problems. In this paper, we provide a survey of RNNs that can be used to solve both the constrained quadratic optimization problem related to support vector machine (SVM) learning, and the SVM model selection...

We propose in this paper a new kernel, suited for Support Vector Machines learning, which is inspired from the biological world. The kernel is based on Gabor filters that are a good model for the response of the cells in the primary visual cortex and have been shown to be very effective in processing natural images. Furthermore, we build a link...

Conventional Machine Learning (ML) algorithms do not contemplate computational constraints when learning models: when targeting their implementation on embedded devices, restrictions are related to, for example, limited depth of the arithmetic unit, memory availability, or battery capacity. We propose a new learning framework, i.e. Algorithmic...

The recurrent network of Xia et al. was proposed for solving quadratic programming problems and was recently adapted to support vector machine (SVM) learning by Tan et al.We show that this formulation contains some unnecessary circuit which, furthermore, can fail to provide the correct value of one of the SVM parameters and suggest how to avoid...

Relevance Vector regression is a form of Support Vector regression, recently proposed by M.E.Tipping, which allows a sparse representation of the data. The Bayesian learning algorithm proposed by the author leaves the partially open question of how to automatically choose the optima! model. In this paper we describe a model selection criterion...

Support Vector Machines are gaining more and more acceptance thanks to their success in many real-world problems. We propose in this work a solution for implementing SVM in hardware. The main idea is to use a recurrent network for SVM learning that guarantees the globally convergence to the optimal solution without the use of penalty terms....

No description

Algorithms, applications and hardware implementations of neural networks are not investigated in close connection. Researchers working in the development of dedicated hardware implementations develop simplified versions of otherwise complex neural algorithms or develop dedicated algorithms: usually these algorithms have not been horoughly...

Streaming Data Analysis (SDA) of Big Data Streams (BDS) for Condition Based Maintenance (CBM) in the context of Rail Transportation Systems (RTS) is a state-of-the-art field of research. SDA of BDS is the problem of analyzing, modeling and extracting information from huge amounts of data that continuously come from several sources in real time...

In this paper we bound the risk of the Gibbs and Bayes classifiers (GC and BC), when the prior is defined in terms of the data generating distribution, and the posterior is defined in terms of the observed one, as proposed by Catoni (2007). We deal with this problem from two different perspectives. From one side we briefly review and further...

No description

In this paper we propose some very simple algorithms and architectures for a digital VLSI implementation of Support Vector Machines. We discuss the main aspects concerning the realization of the learning phase of SVMs, with special attention on the effects of fixed-point math for computing and storing the paraneters of the network. Soime...

Several techniques can be retrieved in literature, which cope with the problem of energy load forecasting in the short-term framework, that is hours up to few days ahead. However, in order to properly schedule operative conditions including e- nergy purchasing and generation, fuel supply, and infrastruc- ture development and maintenance, a...

No description

In this paper we further develop the idea that the PAC-Bayes prior can be defined based on the data-generating distribution. In particular, following Catoni [1], we refine some recent generalisation bounds on the risk of the Gibbs Classifier, when the prior is defined in terms of the data generating distribution, and the posterior is defined in...

No description

Underwater wireless optical communication has been used for establish a link between mobile vehicles and/or fixed nodes because light, especially in the blue/green region, allows to achieve higher data-rate than acoustical or electromagnetic waves for moderate distances. The here proposed work has the aim to pave the way for diffuse optical...

Random Forests (RF) of tree classifiers are a popular ensemble method for classification. RF have shown to be effective in many different real world classification problems and nowadays are considered as one of the best learning algorithms in this context. In this paper we discuss the effect of the hyperparameters of the RF over the accuracy of...

We review in this paper several methods from Statistical Learning Theory (SLT) for the performance assessment and uncertainty quantification of predictive models. Computational issues are addressed so to allow the scaling to large datasets and the application of SLT to Big Data analytics. The effectiveness of the application of SLT to...