State convergence in the initialisation of stream ciphers

An initialisation process is a key component in modern stream cipher design. A well-designed initialisation process should ensure that each key-IV pair generates a different key stream. In this paper, we analyse two ciphers, A5/1 and Mixer, for which this does not happen due to state convergence. We show how the state convergence problem occurs and estimate the effective key-space in each case.

The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network

Sample complexity results from computational learning theory, when applied to neural network learning for pattern classification problems, suggest that for good generalization performance the number of training examples should grow at least linearly with the number of adjustable parameters in the network. Results in this paper show that if a large neural network is used for a pattern classification problem and the learning algorithm finds a network with small weights that has small squared error on the training patterns, then the generalization performance depends on the size of the weights rather than the number of weights. For example, consider a two-layer feedforward network of sigmoid units, in which the sum of the magnitudes of the weights associated with each unit is bounded by A and the input dimension is n. We show that the misclassification probability is no more than a certain error estimate (that is related to squared error on the training set) plus A3 √((log n)/m) (ignoring log A and log m factors), where m is the number of training patterns. This may explain the generalization performance of neural networks, particularly when the number of training examples is considerably smaller than the number of weights. It also supports heuristics (such as weight decay and early stopping) that attempt to keep the weights small during training. The proof techniques appear to be useful for the analysis of other pattern classifiers: when the input domain is a totally bounded metric space, we use the same approach to give upper bounds on misclassification probability for classifiers with decision boundaries that are far from the training examples.

Learning the kernel matrix with semidefinite programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive semidefinite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space - classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semidefinite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -using the labeled part of the data one can learn an embedding also for the unlabeled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method for learning the 2-norm soft margin parameter in support vector machines, solving an important open problem.

Boosting the margin: A new explanation for the effectiveness of voting methods

One of the surprising recurring phenomena observed in experiments with boosting is that the test error of the generated classifier usually does not increase as its size becomes very large, and often is observed to decrease even after the training error reaches zero. In this paper, we show that this phenomenon is related to the distribution of margins of the training examples with respect to the generated voting classification rule, where the margin of an example is simply the difference between the number of correct votes and the maximum number of votes received by any incorrect label. We show that techniques used in the analysis of Vapnik's support vector classifiers and of neural networks with small weights can be applied to voting methods to relate the margin distribution to the test error. We also show theoretically and experimentally that boosting is especially effective at increasing the margins of the training examples. Finally, we compare our explanation to those based on the bias-variance decomposition.

Rademacher and Gaussian complexities: Risk bounds and structural results

We investigate the use of certain data-dependent estimates of the complexity of a function class, called Rademacher and Gaussian complexities. In a decision theoretic setting, we prove general risk bounds in terms of these complexities. We consider function classes that can be expressed as combinations of functions from basis classes and show how the Rademacher and Gaussian complexities of such a function class can be bounded in terms of the complexity of the basis classes. We give examples of the application of these techniques in finding data-dependent risk bounds for decision trees, neural networks and support vector machines.

Local Rademacher complexities

We propose new bounds on the error of learning algorithms in terms of a data-dependent notion of complexity. The estimates we establish give optimal rates and are based on a local and empirical version of Rademacher averages, in the sense that the Rademacher averages are computed from the data, on a subset of functions with small empirical error. We present some applications to classification and prediction with convex function classes, and with kernel classes in particular.

Comment

The support vector machine (SVM) has played an important role in bringing certain themes to the fore in computationally oriented statistics. However, it is important to place the SVM in context as but one member of a class of closely related algorithms for nonlinear classification. As we discuss, several of the “open problems” identified by the authors have in fact been the subject of a significant literature, a literature that may have been missed because it has been aimed not only at the SVM but at a broader family of algorithms. Keeping the broader class of algorithms in mind also helps to make clear that the SVM involves certain specific algorithmic choices, some of which have favorable consequences and others of which have unfavorable consequences—both in theory and in practice. The broader context helps to clarify the ties of the SVM to the surrounding statistical literature.

Adaboost is consistent

The risk, or probability of error, of the classifier produced by the AdaBoost algorithm is investigated. In particular, we consider the stopping strategy to be used in AdaBoost to achieve universal consistency. We show that provided AdaBoost is stopped after n1-ε iterations---for sample size n and ε ∈ (0,1)---the sequence of risks of the classifiers it produces approaches the Bayes risk.

Reverse biased Pt/nanostructured MoO3/SiC Schottky diode based hydrogen gas sensors

Pt/nanostructured molybdenum oxide (MoO3) /SiC Schottky diode based gas sensors were fabricated for hydrogen (H2) gas sensing. Due to the enhanced performance, which is ascribed to the application of MoO3 nanostructures, these devices were used in reversed bias. MoO3 characterization by scanning electron microscopy showed morphology of randomly orientated nanoplatelets with thicknesses between 50 and 500 nm. An α-Β mixed phase crystallographic structure of MoO3 was characterized by x-ray diffraction. At 180 °C, 1.343 V voltage shift in the reverse I-V curve and a Pt/ MoO3 barrier height change of 20 meV were obtained after exposure to 1% H2 gas in synthetic air. © 2009 American Institute of Physics.

A plasmonic 'ac Wheatstone bridge' circuit for high-sensitivity phase measurement and single-molecule detection

In this paper, a plasmonic “ac Wheatstone bridge” circuit is proposed and theoretically modeled for the first time. The bridge circuit consists of three metallic nanoparticles, shaped as rectangular prisms, with two nanoparticles acting as parallel arms of a resonant circuit and the third bridging nanoparticle acting as an optical antenna providing an output signal. Polarized light excites localized surface plasmon resonances in the two arms of the circuit, which generate an optical signal dependent on the phase-sensitive excitations of surface plasmons in the antenna. The circuit is analyzed using a plasmonic coupling theory and numerical simulations. The analyses show that the plasmonic circuit is sensitive to phase shifts between the arms of the bridge and has the potential to detect the presence of single molecules.

Analysis of efficiency and optimization of plasmon energy coupling into nanofocusing metal wedges

In this paper, we investigate theoretically and numerically the efficiency of energy coupling from a plasmon generated by a grating coupler at one of the interfaces of a metal wedge into the plasmonic eigenmode (i.e., symmetric or quasisymmetric plasmon) experiencing nanofocusing in the wedge. Thus the energy efficiency of energy coupling into metallic nanofocusing structure is analyzed. Two different nanofocusing structures with the metal wedge surrounded by a uniform dielectric (symmetric structure) and with the metal wedge enclosed between a substrate and a cladding with different dielectricpermittivities (asymmetric structure) are considered by means of the geometrical optics (adiabatic) approximation. It is demonstrated that the efficiency of the energy coupling from the plasmon generated by the grating into the symmetric or quasisymmetric plasmon experiencing nanofocusing may vary between ∼50% to ∼100%. In particular, even a very small difference (of ∼1%–2%) between the permittivities of the substrate and the cladding may result in a significant increase in the efficiency of the energy coupling (from ∼50% up to ∼100%) into the plasmon experiencing nanofocusing. Distinct beat patterns produced by the interference of the symmetric (quasisymmetric) and antisymmetric (quasiantisymmetric) plasmons are predicted and analyzed with significant oscillations of the magnetic and electric field amplitudes at both the metal wedge interfaces. Physical interpretations of the predicted effects are based upon the behavior, dispersion, and dissipation of the symmetric (quasisymmetric) and antisymmetric (quasiantisymmetric) filmplasmons in the nanofocusing metal wedge. The obtained results will be important for optimizing metallic nanofocusing structures and minimizing coupling and dissipative losses.

A compact interferometric sensor design using three waveguide coupling

The use of metal stripes for the guiding of plasmons is a well established technique for the infrared regime and has resulted in the development of a myriad of passive optical components and sensing devices. However, the plasmons suffer from large losses around sharp bends, making the compact design of nanoscale sensors and circuits problematic. A compact alternative would be to use evanescent coupling between two sufficiently close stripes, and thus we propose a compact interferometer design using evanescent coupling. The sensitivity of the design is compared with that achieved using a hand-held sensor based on the Kretschmann style surface plasmon resonance technique. Modeling of the new interferometric sensor is performed for various structural parameters using finite-difference time-domain and COMSOL Multiphysics. The physical mechanisms behind the coupling and propagation of plasmons in this structure are explained in terms of the allowed modes in each section of the device.

Coherent superposition of exciton states in quantum dots induced by surface plasmons

We present an experimental demonstration of strong optical coupling between CdSequantum dots of different sizes which is induced by a surface plasmon propagating on a planar silver thin film. Attenuated total reflection measurements demonstrate the hybridization of exciton states, characterized by the observation of two avoided crossings in the energy dispersion measured for the interacting system.