Search results for " Neural Networks"

showing 10 items of 390 documents

Human experts vs. machines in taxa recognition

2020

The step of expert taxa recognition currently slows down the response time of many bioassessments. Shifting to quicker and cheaper state-of-the-art machine learning approaches is still met with expert scepticism towards the ability and logic of machines. In our study, we investigate both the differences in accuracy and in the identification logic of taxonomic experts and machines. We propose a systematic approach utilizing deep Convolutional Neural Nets with the transfer learning paradigm and extensively evaluate it over a multi-pose taxonomic dataset with hierarchical labels specifically created for this comparison. We also study the prediction accuracy on different ranks of taxonomic hier…

FOS: Computer and information sciencesComputer Science - Machine Learninghahmontunnistus (tietotekniikka)Computer scienceClassification approachTaxonomic expert02 engineering and technologyneuroverkotcomputer.software_genreConvolutional neural networkQuantitative Biology - Quantitative MethodsField (computer science)Machine Learning (cs.LG)Machine learning approachesStatistics - Machine LearningAutomated approachDeep neural networks0202 electrical engineering electronic engineering information engineeringTaxonomic rankQuantitative Methods (q-bio.QM)Classification (of information)Artificial neural networksystematiikka (biologia)Prediction accuracyIdentification (information)koneoppiminenMulti-image dataBenchmark (computing)020201 artificial intelligence & image processingConvolutional neural networksComputer Vision and Pattern RecognitionClassification errorsMachine Learning (stat.ML)Machine learningState of the artElectrical and Electronic EngineeringTaxonomySupport vector machinesLearning systemsbusiness.industryNode (networking)020206 networking & telecommunicationsComputer circuitsHierarchical classificationConvolutionSupport vector machineFOS: Biological sciencesTaxonomic hierarchySignal ProcessingBiomonitoringBenchmark datasetsArtificial intelligencebusinesscomputertaksonitSoftware
researchProduct

Rule Extraction From Binary Neural Networks With Convolutional Rules for Model Validation.

2020

Classification approaches that allow to extract logical rules such as decision trees are often considered to be more interpretable than neural networks. Also, logical rules are comparatively easy to verify with any possible input. This is an important part in systems that aim to ensure correct operation of a given model. However, for high-dimensional input data such as images, the individual symbols, i.e. pixels, are not easily interpretable. Therefore, rule-based approaches are not typically used for this kind of high-dimensional data. We introduce the concept of first-order convolutional rules, which are logical rules that can be extracted using a convolutional neural network (CNN), and w…

FOS: Computer and information sciencesComputer Science - Machine Learningstochastic local searchrule extractionComputer Science - Artificial Intelligencelogical rulesQA75.5-76.95004 InformatikMachine Learning (cs.LG)Artificial Intelligence (cs.AI)Artificial IntelligenceElectronic computers. Computer scienceconvolutional neural networksk-term DNFinterpretability004 Data processingOriginal ResearchFrontiers in artificial intelligence
researchProduct

Local Granger causality

2021

Granger causality is a statistical notion of causal influence based on prediction via vector autoregression. For Gaussian variables it is equivalent to transfer entropy, an information-theoretic measure of time-directed information transfer between jointly dependent processes. We exploit such equivalence and calculate exactly the 'local Granger causality', i.e. the profile of the information transfer at each discrete time point in Gaussian processes; in this frame Granger causality is the average of its local version. Our approach offers a robust and computationally fast method to follow the information transfer along the time history of linear stochastic processes, as well as of nonlinear …

FOS: Computer and information sciencesInformation transferGaussianFOS: Physical sciencestechniques; information theory; granger causalityMachine Learning (stat.ML)Quantitative Biology - Quantitative Methods01 natural sciences010305 fluids & plasmasVector autoregressionsymbols.namesakegranger causalityGranger causalityStatistics - Machine Learning0103 physical sciencesApplied mathematicstime serie010306 general physicsQuantitative Methods (q-bio.QM)Mathematicsinformation theoryStochastic processDisordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural NetworksComputational Physics (physics.comp-ph)Discrete time and continuous timeAutoregressive modelFOS: Biological sciencesSettore ING-INF/06 - Bioingegneria Elettronica E InformaticasymbolsTransfer entropytechniquesPhysics - Computational Physics
researchProduct

Time Difference of Arrival Estimation from Frequency-Sliding Generalized Cross-Correlations Using Convolutional Neural Networks

2020

The interest in deep learning methods for solving traditional signal processing tasks has been steadily growing in the last years. Time delay estimation (TDE) in adverse scenarios is a challenging problem, where classical approaches based on generalized cross-correlations (GCCs) have been widely used for decades. Recently, the frequency-sliding GCC (FS-GCC) was proposed as a novel technique for TDE based on a sub-band analysis of the cross-power spectrum phase, providing a structured two-dimensional representation of the time delay information contained across different frequency bands. Inspired by deep-learning-based image denoising solutions, we propose in this paper the use of convolutio…

FOS: Computer and information sciencesSound (cs.SD)Computer sciencePhase (waves)Distributed microphones02 engineering and technologyConvolutional neural networkComputer Science - Sound030507 speech-language pathology & audiology03 medical and health sciencesAudio and Speech Processing (eess.AS)FOS: Electrical engineering electronic engineering information engineering0202 electrical engineering electronic engineering information engineeringGCCRepresentation (mathematics)Signal processingbusiness.industryI.5.4Deep learningConvolutional Neural Networks020206 networking & telecommunicationsTime delay estimationMultilaterationI.2.094A12 68T10LocalizationArtificial intelligence0305 other medical sciencebusinessAlgorithmElectrical Engineering and Systems Science - Audio and Speech ProcessingI.2.0; I.5.4ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
researchProduct

Sector identification in a set of stock return time series traded at the London Stock Exchange

2005

We compare some methods recently used in the literature to detect the existence of a certain degree of common behavior of stock returns belonging to the same economic sector. Specifically, we discuss methods based on random matrix theory and hierarchical clustering techniques. We apply these methods to a portfolio of stocks traded at the London Stock Exchange. The investigated time series are recorded both at a daily time horizon and at a 5-minute time horizon. The correlation coefficient matrix is very different at different time horizons confirming that more structured correlation coefficient matrices are observed for long time horizons. All the considered methods are able to detect econo…

FOS: Economics and businessPhysics - Physics and SocietyStatistical Finance (q-fin.ST)SYSTEMSEXPRESSION DATAQuantitative Finance - Statistical FinanceFOS: Physical sciencesFINANCIAL-MARKETSDisordered Systems and Neural Networks (cond-mat.dis-nn)Physics and Society (physics.soc-ph)Condensed Matter - Disordered Systems and Neural NetworksMATRICESNOISE
researchProduct

Orbital Rotations induced by Charges of Polarons and Defects in Doped Vanadates

2020

We explore the competiton of doped holes and defects that leads to the loss of orbital order in vanadate perovskites. In compounds such as La$_{1-{\sf x}}$Ca$_{\,\sf x}$VO$_3$ spin and orbital order result from super-exchange interactions described by an extended three-orbital degenerate Hubbard-Hund model for the vanadium $t_{2g}$ electrons. Long-range Coulomb potentials of charged Ca$^{2+}$ defects and $e$-$e$ interactions control the emergence of defect states inside the Mott gap. The quadrupolar components of the Coulomb fields of doped holes induce anisotropic orbital rotations of degenerate orbitals. These rotations modify the spin-orbital polaron clouds and compete with orbital rotat…

FOS: Physical sciences02 engineering and technologyElectronPolaron01 natural sciencesCondensed Matter - Strongly Correlated ElectronsAtomic orbital0103 physical sciencesCoulomb010306 general physicsSpin (physics)Condensed Matter - Statistical MechanicsPhysicsCondensed Matter - Materials ScienceStrongly Correlated Electrons (cond-mat.str-el)Statistical Mechanics (cond-mat.stat-mech)Condensed matter physicsMaterials Science (cond-mat.mtrl-sci)Order (ring theory)Disordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural Networks021001 nanoscience & nanotechnologySuperexchangeCharge carrierCondensed Matter::Strongly Correlated ElectronsAstrophysics::Earth and Planetary Astrophysics0210 nano-technology
researchProduct

Exact analytic solution of the multi-dimensional Anderson localization

2004

The method proposed by the present authors to deal analytically with the problem of Anderson localization via disorder [J.Phys.: Condens. Matter {\bf 14} (2002) 13777] is generalized for higher spatial dimensions D. In this way the generalized Lyapunov exponents for diagonal correlators of the wave function, $$, can be calculated analytically and exactly. This permits to determine the phase diagram of the system. For all dimensions $D > 2$ one finds intervals in the energy and the disorder where extended and localized states coexist: the metal-insulator transition should thus be interpreted as a first-order transition. The qualitative differences permit to group the systems into two classes…

FOS: Physical sciencesDisordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural Networks
researchProduct

Ar:N$_2$ - a non-universal glass

2014

The bias energies of various two-level systems (TLSs) and their strengths of interactions with the strain are calculated for Ar:N$_2$ glass. Unlike the case in KBr:CN, a distinct class of TLSs having weak interaction with the strain and untypically small bias energies is not found. The addition of CO molecules introduces CO flips which form such a class of weakly interacting TLSs, albeit at much lower coupling than are typically observed in solids. We conclude that because of the absence of a distinct class of weakly interacting TLSs, Ar:N$_2$ is a non-universal glass, the first such system in three dimensions and in ambient pressure. Our results further suggest that Ar:N$_2$:CO may show un…

FOS: Physical sciencesDisordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural Networks
researchProduct

A new approach to the analytic solution of the Anderson localization problem for arbitrary dimensions

2005

Subsequent to the ideas presented in our previous papers [J.Phys.: Condens. Matter {\bf 14} (2002) 13777 and Eur. Phys. J. B {\bf 42} (2004) 529], we discuss here in detail a new analytical approach to calculating the phase-diagram for the Anderson localization in arbitrary spatial dimensions. The transition from delocalized to localized states is treated as a generalized diffusion which manifests itself in the divergence of averages of wavefunctions (correlators). This divergence is controlled by the Lyapunov exponent $\gamma$, which is the inverse of the localization length, $\xi=1/\gamma$. The appearance of the generalized diffusion arises due to the instability of a fundamental mode cor…

FOS: Physical sciencesDisordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural Networks
researchProduct

Defects, Disorder, and Strong Electron Correlations in Orbital Degenerate, Doped Mott Insulators.

2015

We elucidate the effects of defect disorder and $e$-$e$ interaction on the spectral density of the defect states emerging in the Mott-Hubbard gap of doped transition-metal oxides, such as Y$_{1-x}$Ca$_{x}$VO$_{3}$. A soft gap of kinetic origin develops in the defect band and survives defect disorder for $e$-$e$ interaction strengths comparable to the defect potential and hopping integral values above a doping dependent threshold, otherwise only a pseudogap persists. These two regimes naturally emerge in the statistical distribution of gaps among different defect realizations, which turns out to be of Weibull type. Its shape parameter $k$ determines the exponent of the power-law dependence o…

FOS: Physical sciencesGeneral Physics and Astronomylaw.inventionCondensed Matter - Strongly Correlated ElectronsPhysics and Astronomy (all)lawMesoscale and Nanoscale Physics (cond-mat.mes-hall)Spin (physics)Condensed Matter - Statistical MechanicsPhysicsCondensed Matter - Materials ScienceStrongly Correlated Electrons (cond-mat.str-el)Statistical Mechanics (cond-mat.stat-mech)Condensed Matter - Mesoscale and Nanoscale PhysicsCondensed matter physicsMott insulatorDopingDegenerate energy levelsMaterials Science (cond-mat.mtrl-sci)Disordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural NetworksKröger–Vink notationDensity of statesCondensed Matter::Strongly Correlated ElectronsScanning tunneling microscopePseudogapPhysical review letters
researchProduct