Search results for "Neural Networks"
showing 10 items of 599 documents
Stochastic Nonlinear Time Series Forecasting Using Time-Delay Reservoir Computers: Performance and Universality
2014
International audience; Reservoir computing is a recently introduced machine learning paradigm that has already shown excellent performances in the processing of empirical data. We study a particular kind of reservoir computers called time-delay reservoirs that are constructed out of the sampling of the solution of a time-delay diFFerential equation and show their good performance in the forecasting of the conditional covariances associated to multivariate discrete-time nonlinear stochastic processes of VEC-GARCH type as well as in the prediction of factual daily market realized volatilities computed with intraday quotes, using as training input daily log-return series of moderate size. We …
Extended causal modeling to assess Partial Directed Coherence in multiple time series with significant instantaneous interactions.
2010
The Partial Directed Coherence (PDC) and its generalized formulation (gPDC) are popular tools for investigating, in the frequency domain, the concept of Granger causality among multivariate (MV) time series. PDC and gPDC are formalized in terms of the coefficients of an MV autoregressive (MVAR) model which describes only the lagged effects among the time series and forsakes instantaneous effects. However, instantaneous effects are known to affect linear parametric modeling, and are likely to occur in experimental time series. In this study, we investigate the impact on the assessment of frequency domain causality of excluding instantaneous effects from the model underlying PDC evaluation. M…
Massive Lesions Classification using Features based on Morphological Lesion Differences
2007
Purpose of this work is the development of an automatic classification system which could be useful for radiologists in the investigation of breast cancer. The software has been designed in the framework of the MAGIC-5 collaboration. In the automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features based on morphological lesion differences. Some classifiers as a Feed Forward Neural Network, a K-Nearest Neighbours and a Support Vector Machine are used to distinguish the pathological records from the healthy ones. The results obtained in terms of sensiti…
Analysis and simulation of creativity learning by means of artificial neural networks
2007
The paper presents a new neural network approach for analysis and simulation of creative behavior. The used concept of Dynamically Controlled Neural Gas (DyCoNG) entails a combination of Dynamically Controlled Network [Perl, J. (2004a). A neural network approach to movement pattern analysis. Human Movement Science,23, 605-620] and Growing Neural Gas (Fritzke, 1995) by quality neurons. A quality neuron reflects the rareness of a piece of information and therefore can measure the originality of a recorded activity that was assigned to the neuron during the network training. The DyCoNG approach was validated using data from a longitudinal field-based study. The creative behavior of 42 particip…
PNeuro: A scalable energy-efficient programmable hardware accelerator for neural networks
2018
Proceedings of a meeting held 19-23 March 2018, Dresden, Germany; International audience; Artificial intelligence and especially Machine Learning recently gained a lot of interest from the industry. Indeed, new generation of neural networks built with a large number of successive computing layers enables a large amount of new applications and services implemented from smart sensors to data centers. These Deep Neural Networks (DNN) can interpret signals to recognize objects or situations to drive decision processes. However, their integration into embedded systems remains challenging due to their high computing needs. This paper presents PNeuro, a scalable energy-efficient hardware accelerat…
Multitasking associative networks.
2012
We introduce a bipartite, diluted and frustrated, network as a sparse restricted Boltzman machine and we show its thermodynamical equivalence to an associative working memory able to retrieve multiple patterns in parallel without falling into spurious states typical of classical neural networks. We focus on systems processing in parallel a finite (up to logarithmic growth in the volume) amount of patterns, mirroring the low-level storage of standard Amit-Gutfreund-Sompolinsky theory. Results obtained trough statistical mechanics, signal-to-noise technique and Monte Carlo simulations are overall in perfect agreement and carry interesting biological insights. Indeed, these associative network…
Role of noise in a market model with stochastic volatility
2006
We study a generalization of the Heston model, which consists of two coupled stochastic differential equations, one for the stock price and the other one for the volatility. We consider a cubic nonlinearity in the first equation and a correlation between the two Wiener processes, which model the two white noise sources. This model can be useful to describe the market dynamics characterized by different regimes corresponding to normal and extreme days. We analyze the effect of the noise on the statistical properties of the escape time with reference to the noise enhanced stability (NES) phenomenon, that is the noise induced enhancement of the lifetime of a metastable state. We observe NES ef…
Quantification of melanin and hemoglobin in humain skin from multispectral image acquisition: use of a neuronal network combined to a non-negative ma…
2012
International audience; This article presents a multispectral imaging system which, coupled with a neural network-based algorithm, reconstructs reflectance cubes. The reflectance spectra are obtained using artificial neural-netwok reconstruction which generates reflectance cubes from acquired multispectral images. Then, a blind source separation algorithm based on Non-negative Matrix Factorization is used for the decomposition of human skin absorption spectra in its main pigments: melanin and hemoglobin. The analysis is performed on reflectance spectra. The implemented source separation algorithm is based on a multiplicative coefficient upload. The goal is to represent a given spectrum as t…
Advances in photonic reservoir computing
2017
We review a novel paradigm that has emerged in analogue neuromorphic optical computing. The goal is to implement a reservoir computer in optics, where information is encoded in the intensity and phase of the optical field. Reservoir computing is a bio-inspired approach especially suited for processing time-dependent information. The reservoir’s complex and high-dimensional transient response to the input signal is capable of universal computation. The reservoir does not need to be trained, which makes it very well suited for optics. As such, much of the promise of photonic reservoirs lies in their minimal hardware requirements, a tremendous advantage over other hardware-intensive neural net…
A Comparative Analysis of Residual Block Alternatives for End-to-End Audio Classification
2020
Residual learning is known for being a learning framework that facilitates the training of very deep neural networks. Residual blocks or units are made up of a set of stacked layers, where the inputs are added back to their outputs with the aim of creating identity mappings. In practice, such identity mappings are accomplished by means of the so-called skip or shortcut connections. However, multiple implementation alternatives arise with respect to where such skip connections are applied within the set of stacked layers making up a residual block. While residual networks for image classification using convolutional neural networks (CNNs) have been widely discussed in the literature, their a…