On the ERA ranking representability of pairwise bipartite ranking functions

In domains like decision theory and social choice theory it is known for a long time that stochastic transitivity properties yield necessary and sufficient conditions for the ranking or utility representability of reciprocal preference relations. In this article we extend these results for reciprocal preference relations originating from the pairwise comparison of random vectors in a machine learning context. More specifically, the expected ranking accuracy (ERA) is such a reciprocal relation that occurs in multi-class classification problems, when ranking or utility functions are fitted to the data in a pairwise manner. We establish necessary and sufficient conditions for which these pairwise bipartite ranking functions can be simplified to a single ranking function such that the pairwise expected ranking accuracies of both models coincide. Similarly as for more common reciprocal preference relations, cycle transitivity plays a crucial role in this new setting. We first consider the finite sample case, for which expected ranking accuracy can be estimated by means of the area under the ROC curve (AUC), and subsequently, we further generalize these results to the underlying distributions. It turns out that the ranking representability of pairwisely compared random vectors can be expressed elegantly in a distribution-independent way by means of a specific type of cycle transitivity, defined by a conjunctor that is closely related to the algebraic product.     

A sub‐pixel mapping algorithm based on sub‐pixel/pixel spatial attraction models

Soft classification techniques avoid the loss of information characteristic to hard classification techniques when handling mixed pixels. Sub‐pixel mapping is a method incorporating benefits of both hard and soft classification techniques. In this paper an algorithm is developed based on sub‐pixel/pixel attractions. The design of the algorithm is accomplished using artificial imagery but testing is done on artificial as well as real synthetic imagery. The algorithm is evaluated both visually and quantitatively using established classification accuracy indices. The resulting images show increased accuracy when compared to hardened soft classifications.     

Systems-in-foil – Devices,fabrication processes and reliability issues

Systems-in-foil are a new class of electronics in which a full system is integrated into a flexible end product. In this paper, we discuss current research activities and state-of-the-art in this field. Furthermore, some of the associated and expected reliability issues will be addressed on the basis of three examples. As a first example we discuss a flexible large area polymeric organic light-emitting (OLED) device. The reliability targets of these devices require protection against the detrimental influence of water. As a second example, we describe a technology for embedding thinned Si chips between polymeric foils where a careful selection of the base materials is important to account for thermal expansion differences. Finally, as a third example, a novel technology for embedding conductive circuitry in a polymeric foil is discussed in which a good matching of the elastic moduli of the polymeric foil and the embedded circuitry is crucial for the flexibility robustness.     

Focusing Polarization Diversity Grating Couplers in Silicon-on-Insulator

We report on compact focusing polarization diversity grating couplers in silicon-on-insulator, which can be used to overcome the polarization dependence of nanophotonic integrated circuits. The minimum fiber-to-fiber polarization dependent loss is 0.4 dB and the focusing grating couplers are as performant as standard 2-D-grating couplers without focusing. In addition, the focusing property of the gratings results in an 8-fold length reduction of the coupling structure as compared to standard nonfocusing 2-D-grating versions.     

Design and Optimization of Electrically Injected InP-Based Microdisk Lasers Integrated on and Coupled to a SOI Waveguide Circuit

We have performed a numerical study involving the design and optimization of InP-based microdisk lasers integrated on and coupled to a nanophotonic silicon-on-insulator (SOI) waveguide circuit, fabricated through bonding technology. The theoretical model was tested by fitting it to the lasing characteristics obtained for fabricated devices, which we presented previously. A good fit was obtained using parameter values that are consistent with numerical simulation. To obtain optimized laser performance, the composition of the InP-based epitaxial layer structure was optimized to minimize internal optical loss for a structure compatible with efficient current injection. Specific attention was paid to a tunnel-junction based approach. Bending loss was quantified to estimate the minimum microdisk diameter. The coupling between the InP microdisk and Si waveguide was calculated as function of the bonding layer thickness, waveguide offset and waveguide width. To study the lateral injection efficiency, an equivalent electrical network was solved and the voltage-current characteristic was calculated. Based on these results, the dominant device parameters were identified, including microdisk thickness and radius, coupling loss and tunnel-junction p-type doping. These parameters were optimized to obtain maximum wall-plug efficiency, for output powers in the range 1-100 W. The results of this optimization illustrate the potential for substantial improvement in laser performance.     

Nested-Ring Mach–Zehnder Interferometer in Silicon-on-Insulator

For the first time, a nested-ring Mach-Zehnder interferometer (MZI) on silicon-on-insulator is realized using a complementary metal-oxide-semiconductor-based process. In this letter, we verify that the device operates in two modes: the inner-loop resonance dominant mode due to strong build-up inside the inner-ring, and the double-Fano resonances mode due to strong light interaction with the outer loop. The results show that the inner-loop resonance is highly sensitive to the MZI arm imbalance compared to the double-Fano resonance mode. Based on these considerations, we obtain a good fit between theory and experiment.     

Longitudinal analysis of semiconductor lasers with low reflectivity facets

An analysis is made of longitudinal effects in semiconductor lasers with low facet reflectivities. For this purpose, a self-consistent model is used based on the beam propagation method, which takes into account both the lateral and longitudinal dimension. The calculations show that longitudinal effects have a significant influence on the output fields in the laser.     

Experimental results on adiabatic coupling into SOI photonic Crystal coupled-cavity waveguides

The experimental implementation of an adiabatic coupling technique for efficient coupling between photonic crystal single-line defect waveguides and coupled-cavity waveguides is reported. The samples were fabricated in a silicon-on-insulator substrate using 248-nm-deep ultraviolet lithography. Experimental results are compared with three-dimensional (3-D) finite-difference time-domain simulations. Furthermore, the discrepancies between two-dimensional and 3-D simulation results are analyzed.     

Improvement of the wavelength switching behavior of semiconductor tunable lasers through optical feedback from a periodic reference filter based on a novel prism-based implementation of a Fox-Smith resonator

We show that the wavelength switching behavior of semiconductor tunable lasers can be improved through optical feedback from a stable reference filter. The filter is based on a novel prism-based implementation of a Fox-Smith resonator and has a response consisting of periodically spaced peaks, both in reflection and transmission. The improvement of the wavelength switching behavior stems from the suppression of the thermally induced wavelength drift associated with the switching of the tuning currents.     

VH gene family utilization of anti-acetylcholine receptor antibodies in experimental autoimmune myasthenia gravis

The immunoglobulin heavy chain (VH) gene family usage in experimental autoimmune myasthenia gravis (EAMG) model was investigated by RNA slot blot hybridization using VH gene family specific probes. Anti-acetylcholine receptor (AChR) monoclonal antibodies (mAbs) isolated from susceptible C57BL/6 and resistant BALB/c mice were found to be encoded by VH genes from at least six different families. The Vgam3.8 family was overrepresented in alpha-bungarotoxin blocking mAbs. Expression of cross-reactive idiotypes by anti-AChR mAbs was irrespective of the VH gene family usage. Strain dependent differences in susceptibility for EAMG were not reflected in an aberrant VH gene family usage of anti-AChR mAbs.