Globally optimal on-line learning rules for multi-layer neural networks

We present a method for determining the globally optimal on-line learning rule for a soft committee machine under a statistical mechanics framework. This rule maximizes the total reduction in generalization error over the whole learning process. A simple example demonstrates that the locally optimal rule, which maximizes the rate of decrease in generalization error, may perform poorly in comparison.

Efficient THz radiation from a nanocrystalline silicon-based multi-layer photomixer

In this paper we propose a novel type of multiple-layer photomixer based on amorphous/nano-crystalline-Si. Such a device implies that it could be possible to enhance the conversion efficiency from optical power to THz emission by increasing the absorption length and by reducing the device overheating through the use of substrates with higher thermal conductivity compared to GaAs. Our calculations show that the output power from a two-layer Si-based photomixer is at least ten times higher than that from conventional LT-GaAs photomixers at 1 THz.

Physical characteristics of localized surface plasmons resulting from nano-scale structured multi-layer thin films deposited on D-shaped optical fiber

Novel surface plasmonic optical fiber sensors have been fabricated using multiple coatings deposited on a lapped section of a single mode fiber. UV laser irradiation processing with a phase mask produces a nano-scaled surface relief grating structure resembling nano-wires. The resulting individual corrugations produced by material compaction are approximately 20 μm long with an average width at half maximum of 100 nm and generate localized surface plasmons. Experimental data are presented that show changes in the spectral characteristics after UV processing, coupled with an overall increase in the sensitivity of the devices to surrounding refractive index. Evidence is presented that there is an optimum UV dosage (48 joules) over which no significant additional optical change is observed. The devices are characterized with regards to change in refractive index, where significantly high spectral sensitivities in the aqueous index regime are found, ranging up to 4000 nm/RIU for wavelength and 800 dB/RIU for intensity. © 2013 Optical Society of America.

On the k-anonymization of time-varying and multi-layer social graphs

The popularity of online social media platforms provides an unprecedented opportunity to study real-world complex networks of interactions. However, releasing this data to researchers and the public comes at the cost of potentially exposing private and sensitive user information. It has been shown that a naive anonymization of a network by removing the identity of the nodes is not sufficient to preserve users’ privacy. In order to deal with malicious attacks, k -anonymity solutions have been proposed to partially obfuscate topological information that can be used to infer nodes’ identity. In this paper, we study the problem of ensuring k anonymity in time-varying graphs, i.e., graphs with a structure that changes over time, and multi-layer graphs, i.e., graphs with multiple types of links. More specifically, we examine the case in which the attacker has access to the degree of the nodes. The goal is to generate a new graph where, given the degree of a node in each (temporal) layer of the graph, such a node remains indistinguishable from other k-1 nodes in the graph. In order to achieve this, we find the optimal partitioning of the graph nodes such that the cost of anonymizing the degree information within each group is minimum. We show that this reduces to a special case of a Generalized Assignment Problem, and we propose a simple yet effective algorithm to solve it. Finally, we introduce an iterated linear programming approach to enforce the realizability of the anonymized degree sequences. The efficacy of the method is assessed through an extensive set of experiments on synthetic and real-world graphs.

Towards high performance and multi-functional structural membranes using advanced fibrous and textile materials

Scientific and technological advancements in the area of fibrous and textile materials have greatly enhanced their application potential in several high-end technical and industrial sectors including construction, transportation, medical, sports, aerospace engineering, electronics and so on. Excellent performance accompanied by light-weight, mechanical flexibility, tailor-ability, design flexibility, easy fabrication and relatively lower cost are the driving forces towards wide applications of these materials. Cost-effective fabrication of various advanced and functional materials for structural parts, medical devices, sensors, energy harvesting devices, capacitors, batteries, and many others has been possible using fibrous and textile materials. Structural membranes are one of the innovative applications of textile structures and these novel building skins are becoming very popular due to flexible design aesthetics, durability, lightweight and cost benefits. Current demand on high performance and multi-functional materials in structural applications has motivated to go beyond the basic textile structures used for structural membranes and to use innovative textile materials. Structural membranes with self-cleaning, thermoregulation and energy harvesting capability (using solar cells) are examples of such recently developed multi-functional membranes. Besides these, there exist enormous opportunities to develop wide varieties of multi-functional membranes using functional textile materials. Additionally, it is also possible to further enhance the performance and functionalities of structural membranes using advanced fibrous architectures such as 2D, 3D, hybrid, multi-layer and so on. In this context, the present paper gives an overview of various advanced and functional fibrous and textile materials which have enormous application potential in structural membranes.

A Multi-scale Model for Copper Dishing in Chemical-Mechanical Polishing

The present success in the manufacture of multi-layer interconnects in ultra-large-scale integration is largely due to the acceptable planarization capabilities of the chemical-mechanical polishing (CMP) process. In the past decade, copper has emerged as the preferred interconnect material. The greatest challenge in Cu CMP at present is the control of wafer surface non-uniformity at various scales. As the size of a wafer has increased to 300 mm, the wafer-level non-uniformity has assumed critical importance. Moreover, the pattern geometry in each die has become quite complex due to a wide range of feature sizes and multi-level structures. Therefore, it is important to develop a non-uniformity model that integrates wafer-, die- and feature-level variations into a unified, multi-scale dielectric erosion and Cu dishing model. In this paper, a systematic way of characterizing and modeling dishing in the single-step Cu CMP process is presented. The possible causes of dishing at each scale are identified in terms of several geometric and process parameters. The feature-scale pressure calculation based on the step-height at each polishing stage is introduced. The dishing model is based on pad elastic deformation and the evolving pattern geometry, and is integrated with the wafer- and die-level variations. Experimental and analytical means of determining the model parameters are outlined and the model is validated by polishing experiments on patterned wafers. Finally, practical approaches for minimizing Cu dishing are suggested.

A Multi-scale Model for Copper Dishing in Chemical-Mechanical Polishing

The present success in the manufacture of multi-layer interconnects in ultra-large-scale integration is largely due to the acceptable planarization capabilities of the chemical-mechanical polishing (CMP) process. In the past decade, copper has emerged as the preferred interconnect material. The greatest challenge in Cu CMP at present is the control of wafer surface non-uniformity at various scales. As the size of a wafer has increased to 300 mm, the wafer-level non-uniformity has assumed critical importance. Moreover, the pattern geometry in each die has become quite complex due to a wide range of feature sizes and multi-level structures. Therefore, it is important to develop a non-uniformity model that integrates wafer-, die- and feature-level variations into a unified, multi-scale dielectric erosion and Cu dishing model. In this paper, a systematic way of characterizing and modeling dishing in the single-step Cu CMP process is presented. The possible causes of dishing at each scale are identified in terms of several geometric and process parameters. The feature-scale pressure calculation based on the step-height at each polishing stage is introduced. The dishing model is based on pad elastic deformation and the evolving pattern geometry, and is integrated with the wafer- and die-level variations. Experimental and analytical means of determining the model parameters are outlined and the model is validated by polishing experiments on patterned wafers. Finally, practical approaches for minimizing Cu dishing are suggested.

Spectroscopic evidence of extended states in quantized Hall phase of weakly coupled GaAs/AlGaAs multi-layers

The influence of the interlayer coupling on formation of the quantized Hall conductor phase at the filling factor v = 2 was studied in the multi-layer GaAs/AlGaAs heterostructures. The disorder broadened Gaussian photoluminescence line due to the localized electrons was found in the quantized Hall phase of the isolated multi-quantum well structure. On the other hand, the quantized Hall phase of the weakly coupled multi-layers emitted an unexpected asymmetrical line similar to that one observed in the metallic electron systems. We demonstrated that the observed asymmetry is caused by a partial population of the extended electron states formed in the quantized Hall conductor phase due to the interlayer percolation. A sharp decrease of the single-particle scattering time associated with these extended states was observed at the filling factor v = 2. (c) 2007 Elsevier B.V. All rights reserved.

Burning Graphene Layer-by-Layer

Graphene, in single layer or multi-layer forms, holds great promise for future electronics and high-temperature applications. Resistance to oxidation, an important property for high-temperature applications, has not yet been extensively investigated. Controlled thinning of multi-layer graphene (MLG), e.g., by plasma or laser processing is another challenge, since the existing methods produce non-uniform thinning or introduce undesirable defects in the basal plane. We report here that heating to extremely high temperatures (exceeding 2000 K) and controllable layer-by-layer burning (thinning) can be achieved by low-power laser processing of suspended high-quality MLG in air in "cold-wall" reactor configuration. In contrast, localized laser heating of supported samples results in non-uniform graphene burning at much higher rates. Fully atomistic molecular dynamics simulations were also performed to reveal details of oxidation mechanisms leading to uniform layer-by-layer graphene gasification. The extraordinary resistance of MLG to oxidation paves the way to novel high-temperature applications as continuum light source or scaffolding material.

Hierarchical multi-label classification using local neural networks

Hierarchical multi-label classification is a complex classification task where the classes involved in the problem are hierarchically structured and each example may simultaneously belong to more than one class in each hierarchical level. In this paper, we extend our previous works, where we investigated a new local-based classification method that incrementally trains a multi-layer perceptron for each level of the classification hierarchy. Predictions made by a neural network in a given level are used as inputs to the neural network responsible for the prediction in the next level. We compare the proposed method with one state-of-the-art decision-tree induction method and two decision-tree induction methods, using several hierarchical multi-label classification datasets. We perform a thorough experimental analysis, showing that our method obtains competitive results to a robust global method regarding both precision and recall evaluation measures.

Evaluation of a multi variated analysis for the selection of t-tbar multijet events in the current CMS implemented environment at LHC

In the present study we are using multi variate analysis techniques to discriminate signal from background in the fully hadronic decay channel of ttbar events. We give a brief introduction to the role of the Top quark in the standard model and a general description of the CMS Experiment at LHC. We have used the CMS experiment computing and software infrastructure to generate and prepare the data samples used in this analysis. We tested the performance of three different classifiers applied to our data samples and used the selection obtained with the Multi Layer Perceptron classifier to give an estimation of the statistical and systematical uncertainty on the cross section measurement.

Modelli e linguaggi per grafi multi-livello

I grafi sono molto utilizzati per la rappresentazione di dati, sopratutto in quelle aree dove l’informazione sull’interconnettività e la topologia dei dati è importante tanto quanto i dati stessi, se non addirittura di più. Ogni area di applicazione ha delle proprie necessità, sia in termini del modello che rappresenta i dati, sia in termini del linguaggio capace di fornire la necessaria espressività per poter fare interrogazione e trasformazione dei dati. È sempre più frequente che si richieda di analizzare dati provenienti da diversi sistemi, oppure che si richieda di analizzare caratteristiche dello stesso sistema osservandolo a granularità differenti, in tempi differenti oppure considerando relazioni differenti. Il nostro scopo è stato quindi quello di creare un modello, che riesca a rappresentare in maniera semplice ed efficace i dati, in tutte queste situazioni. Entrando più nei dettagli, il modello permette non solo di analizzare la singola rete, ma di analizzare più reti, relazionandole tra loro. Il nostro scopo si è anche esteso nel definire un’algebra, che, tramite ai suoi operatori, permette di compiere delle interrogazioni su questo modello. La definizione del modello e degli operatori sono stati maggiormente guidati dal caso di studio dei social network, non tralasciando comunque di rimanere generali per fare altri tipi di analisi. In seguito abbiamo approfondito lo studio degli operatori, individuando delle proprietà utili per fare delle ottimizzazioni, ragionando sui dettagli implementativi, e fornendo degli algoritmi di alto livello. Per rendere più concreta la definizione del modello e degli operatori, in modo da non lasciare spazio ad ambiguità, è stata fatta anche un’implementazione, e in questo elaborato ne forniremo la descrizione.

Adsorption of N-2, Ar, acetone, chloroform and acetone-chloroform mixture on carbon black in the framework of molecular layer structure theory (MLST)

Adsorption of pure nitrogen, argon, acetone, chloroform and acetone-chloroform mixture on graphitized thermal carbon black is considered at sub-critical conditions by means of molecular layer structure theory (MLST). In the present version of the MLST an adsorbed fluid is considered as a sequence of 2D molecular layers, whose Helmholtz free energies are obtained directly from the analysis of experimental adsorption isotherm of pure components. The interaction of the nearest layers is accounted for in the framework of mean field approximation. This approach allows quantitative correlating of experimental nitrogen and argon adsorption isotherm both in the monolayer region and in the range of multi-layer coverage up to 10 molecular layers. In the case of acetone and chloroform the approach also leads to excellent quantitative correlation of adsorption isotherms, while molecular approaches such as the non-local density functional theory (NLDFT) fail to describe those isotherms. We extend our new method to calculate the Helmholtz free energy of an adsorbed mixture using a simple mixing rule, and this allows us to predict mixture adsorption isotherms from pure component adsorption isotherms. The approach, which accounts for the difference in composition in different molecular layers, is tested against the experimental data of acetone-chloroform mixture (non-ideal mixture) adsorption on graphitized thermal carbon black at 50 degrees C. (C) 2005 Elsevier Ltd. All rights reserved.

Évaluation de la modélisation de la taille de grain de neige du modèle multi-couches thermodynamique SNOWPACK: implication dans l'évaluation des risques d'avalanches

Résumé: L’Institut pour l'étude de la neige et des avalanches en Suisse (SLF) a développé SNOWPACK, un modèle thermodynamique multi-couches de neige permettant de simuler les propriétés géophysiques du manteau neigeux (densité, température, taille de grain, teneur en eau, etc.) à partir desquelles un indice de stabilité est calculé. Il a été démontré qu’un ajustement de la microstructure serait nécessaire pour une implantation au Canada. L'objectif principal de la présente étude est de permettre au modèle SNOWPACK de modéliser de manière plus réaliste la taille de grain de neige et ainsi obtenir une prédiction plus précise de la stabilité du manteau neigeux à l’aide de l’indice basé sur la taille de grain, le Structural Stability Index (SSI). Pour ce faire, l’erreur modélisée (biais) par le modèle a été analysée à l’aide de données précises sur le terrain de la taille de grain à l’aide de l’instrument IRIS (InfraRed Integrated Sphere). Les données ont été recueillies durant l’hiver 2014 à deux sites différents au Canada : parc National des Glaciers, en Colombie-Britannique ainsi qu’au parc National de Jasper. Le site de Fidelity était généralement soumis à un métamorphisme à l'équilibre tandis que celui de Jasper à un métamorphisme cinétique plus prononcé. Sur chacun des sites, la stratigraphie des profils de densités ainsi des profils de taille de grain (IRIS) ont été complétés. Les profils de Fidelity ont été complétés avec des mesures de micropénétromètre (SMP). L’analyse des profils de densité a démontré une bonne concordance avec les densités modélisées (R[indice supérieur 2]=0.76) et donc la résistance simulée pour le SSI a été jugée adéquate. Les couches d’instabilités prédites par SNOWPACK ont été identifiées à l’aide de la variation de la résistance dans les mesures de SMP. L’analyse de la taille de grain optique a révélé une surestimation systématique du modèle ce qui est en accord avec la littérature. L’erreur de taille de grain optique dans un environnement à l’équilibre était assez constante tandis que l’erreur en milieux cinétique était plus variable. Finalement, une approche orientée sur le type de climat représenterait le meilleur moyen pour effectuer une correction de la taille de grain pour une évaluation de la stabilité au Canada.

Development and evaluation of neural network freeway incident detection models using field data

This paper discusses a multi-layer feedforward (MLF) neural network incident detection model that was developed and evaluated using field data. In contrast to published neural network incident detection models which relied on simulated or limited field data for model development and testing, the model described in this paper was trained and tested on a real-world data set of 100 incidents. The model uses speed, flow and occupancy data measured at dual stations, averaged across all lanes and only from time interval t. The off-line performance of the model is reported under both incident and non-incident conditions. The incident detection performance of the model is reported based on a validation-test data set of 40 incidents that were independent of the 60 incidents used for training. The false alarm rates of the model are evaluated based on non-incident data that were collected from a freeway section which was video-taped for a period of 33 days. A comparative evaluation between the neural network model and the incident detection model in operation on Melbourne's freeways is also presented. The results of the comparative performance evaluation clearly demonstrate the substantial improvement in incident detection performance obtained by the neural network model. The paper also presents additional results that demonstrate how improvements in model performance can be achieved using variable decision thresholds. Finally, the model's fault-tolerance under conditions of corrupt or missing data is investigated and the impact of loop detector failure/malfunction on the performance of the trained model is evaluated and discussed. The results presented in this paper provide a comprehensive evaluation of the developed model and confirm that neural network models can provide fast and reliable incident detection on freeways. (C) 1997 Elsevier Science Ltd. All rights reserved.