Aurelio Uncini's Publications List

Flexible nonlinear blind signal separation in the complex domain

2009-06-24T11:06:03Z

This paper introduces an Independent Component Analysis (ICA) approach to the separation of nonlinear mixtures in the complex domain. Source separation is performed by a complex INFOMAX approach. The neural network which realizes the separation employs the so called “Mirror Model” and is based on adaptive activation functions, whose shape is properly modified during learning. Nonlinear functio...

Daniele Vigliano, Michele Scarpiniti, Raffaele Parisi, Aurelio Uncini (2008) INTERNATIONAL JOURNAL OF NEURAL SYSTEMS 18: 2 105-122

Flexible nonlinear blind signal separation in the complex domain.

2009-06-24T09:47:09Z

This paper introduces an Independent Component Analysis (ICA) approach to the separation of nonlinear mixtures in the complex domain. Source separation is performed by a complex INFOMAX approach. The neural network which realizes the separation employs the so called "Mirror Model" and is based on adaptive activation functions, whose shape is properly modified during learning. Nonlinear functions i...

Daniele Vigliano, Michele Scarpiniti, Raffaele Parisi, Aurelio Uncini (2008) Int J Neural Syst 18: 2 105-122

Generalized splitting functions for blind separation of complex signals

2009-06-24T11:06:03Z

This paper proposes the blind separation of complex signals using a novel neural network architecture based on an adaptive nonlinear bi-dimensional activation function (AF); the separation is obtained maximizing the output joint entropy. Avoiding the restriction due to the Louiville’s theorem, the AF is composed of a couple of bi-dimensional spline functions, one for the real and one for the ima...

Michele Scarpiniti, Daniele Vigliano, Raffaele Parisi, Aurelio Uncini (2008) NEUROCOMPUTING 71: 10-12 2245-2270

An information theoretic approach to a novel nonlinear independent component analysis paradigm

2009-06-24T11:06:03Z

This paper introduces a novel independent component analysis (ICA) approach to the separation of nonlinear convolutive mixtures. The proposed model is an extension of the well-known post nonlinear (PNL) mixing model and consists of the convolutive mixing of PNL mixtures. Theoretical proof of existence and uniqueness of the solution under proper assumptions is provided. Feedforward and recurrent de...

D Vigliano, R Parisi, A Uncini (2005) SIGNAL PROCESSING 85: 5 997-1028

Regularising neural networks using flexible multivariate activation function

2009-06-24T11:06:03Z

This paper presents a new general neural structure based on nonlinear flexible multivariate function that can be viewed in the framework of the generalised regularisation net-works theory. The proposed architecture is based on multi-dimensional adaptive cubic spline basis activation function that collects information from the previous network layer in aggregate form. In other words, each activatio...

M Solazzi, A Uncini (2004) NEURAL NETWORKS 17: 2 247-260

Spline neural networks for blind separation of post-nonlinear-linear mixtures

2009-06-24T11:06:03Z

In this paper, a novel paradigm for blind source separation in the presence of nonlinear mixtures is presented. In particular the paper addresses the problem of post-nonlinear mixing followed by another instantaneous mixing system. This model is called here the post-nonlinear-linear model. The method is based on the use of the recently introduced flexible. activation function whose control points ...

M Solazzi, A Uncini (2004) IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 51: 4 817-829

‘Mirror model’ gives separation of convolutive mixing of PNL mixtures

2009-06-24T11:06:03Z

The proof is given that the so called ‘mirror model’ as demixing model is able to recover original sources after non-trivial mixing. The issue explored is the capability to separate sources, in a blind way, after the convolutive mixing of post nonlinear (PNL) mixtures. The strictness of that kind of mixture produces non-trivial problems in separating signals without any adequate assumption on ...

D Vigliano, A Uncini (2004) ELECTRONICS LETTERS 40: 7 454-456

Regularising neural networks using flexible multivariate activation function.

2009-06-24T09:47:09Z

This paper presents a new general neural structure based on nonlinear flexible multivariate function that can be viewed in the framework of the generalised regularisation networks theory. The proposed architecture is based on multi-dimensional adaptive cubic spline basis activation function that collects information from the previous network layer in aggregate form. In other words, each activation...

Mirko Solazzi, Aurelio Uncini (2004) Neural Netw 17: 2 247-260

Blind signal processing by complex domain adaptive spline neural networks

2009-06-24T11:06:03Z

In this paper, neural networks based on an adaptive nonlinear function suitable for both blind complex time domain signal separation and blind frequency domain signal deconvolution, are presented. This activation function, whose shape is modified during learning, is based on a couple of spline functions, one for the real and one for the imaginary part of the input. The shape control points are ada...

A Uncini, F Piazza (2003) IEEE TRANSACTIONS ON NEURAL NETWORKS 14: 2 399-412

Audio signal processing by neural networks

2009-06-24T11:06:03Z

In this paper a review of architectures suitable for nonlinear real-time audio signal processing is presented. The computational and structural complexity of neural networks (NNs) represent in fact, the main drawbacks that can hinder many practical NNs multimedia applications. In particular efficient neural architectures and their learning algorithm for real-time on-line audio processing are discu...

A Uncini (2003) NEUROCOMPUTING 55: 3-4 593-625

Special issue on evolving solution with neural networks

2009-06-24T11:06:03Z

A Fanni, A Uncini (2003) NEUROCOMPUTING 55: 3-4 417-419

Real-time room acoustic response simulation by IIR adaptive filter

2009-06-24T11:06:03Z

A new IIR adaptive filter for real-time, room acoustic response simulation is proposed, the structure of which derives from Jot’s model of an artificial reverberator. The simultaneous perturbation stochastic approximation (SPSA) algorithm is used to set parameter values. Results show good similarity between the desired and artificial response.

G Costantini, A Uncini (2003) ELECTRONICS LETTERS 39: 3 330-332

Blind signal processing by complex domain adaptive spline neural networks.

2009-06-24T09:50:42Z

Flexible ICA solution for nonlinear blind source separation problem

2009-06-24T11:06:03Z

Presented is a new architecture and a new learning algorithm that are exploited to resolve the blind source separation problem under stricter constraints than those considered to date. The mixing model that is assumed is an evolution of the well-known post-nonlinear (PNL) one: the PNL mixing block is followed by a convolutive mixing channel. The flexibility of the algorithm originates from the spl...

D Vigliano, A Uncini (2003) ELECTRONICS LETTERS 39: 22 1616-1617

Learning of physical-like sound synthesis models by adaptive spline recurrent neural networks

2009-06-24T11:06:03Z

A recently introduced neural networks architecture, ‘adaptive spline neural networks’ with FIR/IIR synapse, is used to define a general class of physical-like sound synthesis model. To reduce computational cost, use is made of power-of-two synapses followed by a CR-spline-based flexible activation function the shape of which can be modified through its control points. The learning phase is per...

F Iannelli, A Uncini (2002) ELECTRONICS LETTERS 38: 14 724-725

Subband neural networks prediction for on-line audio signal recovery.

2009-06-24T09:50:42Z

In this paper, a subbands multirate architecture is presented for audio signal recovery. Audio signal recovery is a common problem in digital music signal restoration field, because of corrupted samples that must be replaced. The subband approach allows for the reconstruction of a long audio data sequence from forward-backward predicted samples. In order to improve prediction performances, neural ...

G Cocchi, A Uncini (2002) IEEE Trans Neural Netw 13: 4 867-876

Subband neural networks prediction for on-line audio signal recovery

2009-06-24T11:06:03Z

In this paper, a subbands multirate architecture is presented for audio signal recovery. Audio signal recovery is a common problem in digital music signal restoration field, because of corrupted samples that must be replaced. The sub band approach allows for the reconstruction of along audio data sequence from forward-backward predicted samples. In order to improve prediction performances, neural ...

G Cocchi, A Uncini (2002) IEEE TRANSACTIONS ON NEURAL NETWORKS 13: 4 867-876

Complex discriminative learning Bayesian neural equalizer

2009-06-24T11:06:03Z

Traditional approaches to channel equalization are based on the inversion of the global (linear or nonlinear) channel response. However, in digital links the complete channel inversion is neither required nor desirable. Since transmitted symbols belong to a discrete alphabet, symbol demodulation can be effectively recasted as a classification problem in the space of received symbols. In this paper...

M Solazzi, A Uncini, E D Di Claudio, R Parisi (2001) SIGNAL PROCESSING 81: 12 2493-2502

A signal-flow-graph approach to on-line gradient calculation

2009-06-24T11:06:03Z

A large class of nonlinear dynamic adaptive systems such as dynamic recurrent neural networks can be effectively represented by signal flow graphs (SFGs). By this method, complex systems are described as a general connection of many simple components, each of them implementing a simple one-input, one-output transformation, as in an electrical circuit. Even if graph representations are popular in t...

P Campolucci, A Uncini, F Piazza (2000) NEURAL COMPUTATION 12: 8 1901-1927

Power-of-two adaptive filters using Tabu Search

2009-06-24T11:06:03Z

Digital filters with power-of-two or a sum of power-of-two coefficients can be built using simple and fast shift registers instead of slower floating-point multipliers, such a strategy can reduce both the VLSI silicon area and the computational time. Due to the quantization and the nonuniform distribution of the coefficients through their domain, in the case of adaptive filters, classical steepest...

S Traferro, A Uncini (2000) IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-ANALOG AND DIGITAL SIGNAL PROCESSING 47: 6 566-569

Multilayer feedforward networks with adaptive spline activation function.

2009-06-24T09:50:42Z

In this paper, a new adaptive spline activation function neural network (ASNN) is presented. Due to the ASNN's high representation capabilities, networks with a small number of interconnections can be trained to solve both pattern recognition and data processing real-time problems. The main idea is to use a Catmull-Rom cubic spline as the neuron's activation function, which ensures a simple struct...

S Guarnieri, F Piazza, A Uncini (1999) IEEE Trans Neural Netw 10: 3 672-683

On-line learning algorithms for locally recurrent neural networks.

2009-06-24T09:50:42Z

This paper focuses on online learning procedures for locally recurrent neural nets with emphasis on multilayer perceptron (MLP) with infinite impulse response (IIR) synapses and its variations which include generalized output and activation feedback multilayer networks (MLN). We propose a new gradient-based procedure called recursive backpropagation (RBP) whose online version, causal recursive bac...

P Campolucci, A Uncini, F Piazza, B D Rao (1999) IEEE Trans Neural Netw 10: 2 253-271

Complex-valued neural networks with adaptive spline activation function for digital radio links nonlinear equalization

2009-06-24T11:06:03Z

In this paper, a new complex-valued neural network based on adaptive activation functions is proposed. By varying the control points of a pair of Catmull-Rom cubic splines, which are used as an adaptable activation function, this new kind of neural network can be implemented as a very simple structure that is able to improve the generalization capabilities using few training samples, Due to its lo...

A Uncini, L Vecci, P Campolucci, F Piazza (1999) IEEE TRANSACTIONS ON SIGNAL PROCESSING 47: 2 505-514

On-line learning algorithms for locally recurrent neural networks

2009-06-24T11:06:03Z

This paper focuses on on-line learning procedures for locally recurrent neural networks with emphasis on multilayer perceptron (MLP) with infinite impulse response (IIR) synapses and its variations which include generalized output and activation feedback multilayer networks (MLN’s). We propose a new gradient-based procedure called recursive backpropagation (RBP) whose on-line version, causal rec...

P Campolucci, A Uncini, F Piazza, B D Rao (1999) IEEE TRANSACTIONS ON NEURAL NETWORKS 10: 2 253-271

Multilayer feedforward networks with adaptive spline activation function

2009-06-24T11:06:03Z

in this paper, a new adaptive spline activation function neural network (ASNN) is presented. Due to the ASNN’s high representation capabilities, networks with a small number of interconnections can be trained to solve both pattern recognition and data processing real-time problems. The main idea is to use a Catmull-Rom cubic spline as the neuron’s activation function, which ensures a simple st...

S Guarnieri, F Piazza, A Uncini (1999) IEEE TRANSACTIONS ON NEURAL NETWORKS 10: 3 672-683

Learning and Approximation Capabilities of Adaptive Spline Activation Function Neural Networks.

2009-06-24T09:47:09Z

In this paper, we study the theoretical properties of a new kind of artificial neural network, which is able to adapt its activation functions by varying the control points of a Catmull-Rom cubic spline. Most of all, we are interested in generalization capability, and we can show that our architecture presents several advantages. First of all, it can be seen as a sub-optimal realization of the add...

Uncini, Piazza, Vecci (1998) Neural Netw 11: 2 259-270

Learning and approximation capabilities of adaptive spline activation function neural networks

2009-06-24T11:06:03Z

In this paper, we study the theoretical properties of a new kind of artificial neural network, which is able to adapt its activation functions by varying the control points of a Catmull-Rom cubic spline. Most of all, we are interested in generalization capability, and we can show that our architecture presents several advantages. First of all, it can be seen as a sub-optimal realization of the add...

L Vecci, F Piazza, A Uncini (1998) NEURAL NETWORKS 11: 2 259-270

Generalised backpropagation algorithm for training a data predistorter with memory in radio systems

2009-06-24T11:06:03Z

The authors present a neural network based data-predistorter with memory, for the compensation of high-power amplifier (HPA) nonlinearities in digital microwave radio systems. The overall system (predistorter, pulse shaping filter and HPA) can be seen as a unique FIR multilayer neural network, for which a specific complex-valued back-propagation algorithm can be developed to realise the data predi...

N Benvenuto, F Piazza, A Uncini, M Visintin (1996) ELECTRONICS LETTERS 32: 20 1925-1926

Backpropagation without multiplier for multilayer neural networks

2009-06-24T11:06:03Z

When multilayer neural networks are implemented with digital hardware, which allows full exploitation of the well developed digital VLSI technologies, the multiply operations in each neuron between the weights and the inputs can create a bottleneck in the system, because the digital multipliers are very demanding in terms of time or chip area. For this reason, the use of weights constrained to be ...

M L Marchesi, F Piazza, A Uncini (1996) IEE PROCEEDINGS-CIRCUITS DEVICES AND SYSTEMS 143: 4 229-232

FAST NEURAL NETWORKS WITHOUT MULTIPLIERS

2009-06-24T11:06:03Z

The paper introduces multilayer perceptrons with weight values restricted to powers-of-two or sum of power-of-two. In a digital implementation, these neural networks do not need multipliers but only shift registers when computing in forward mode, thus saving chip area and computation time. A learning procedure, based on back-propagation, is presented for such neural networks. This learning procedu...

M MARCHESI, G ORLANDI, F PIAZZA, A UNCINI (1993) IEEE TRANSACTIONS ON NEURAL NETWORKS 4: 1 53-62

APPLICATIONS OF SIMULATED ANNEALING FOR THE DESIGN OF SPECIAL DIGITAL-FILTERS

2009-06-24T11:06:03Z

This paper describes the salient features of using a simulated annealing (SA) algorithm in the context of designing digital filters with coefficient values expressed as the sum of power of two. A procedure for linear phase digital filter design, using this algorithm, is first presented and tested, yielding results as good as known optimal methods. The algorithm is then applied to the design of Nyq...

N BENVENUTO, M MARCHESI, A UNCINI (1992) IEEE TRANSACTIONS ON SIGNAL PROCESSING 40: 2 323-332