Data-dependent dissimilarity measure: an effective alternative to geometric distance measures

Nearest neighbor search is a core process in many data mining algorithms. Finding reliable closest matches of a test instance is still a challenging task as the effectiveness of many general-purpose distance measures such as \(\ell _p\)-norm decreases as the number of dimensions increases. Their performances vary significantly in different data distributions. This is mainly because they compute the distance between two instances solely based on their geometric positions in the feature space, and data distribution has no influence on the distance measure. This paper presents a simple data-dependent general-purpose dissimilarity measure called ‘\(m_p\)-dissimilarity’. Rather than relying on geometric distance, it measures the dissimilarity between two instances as a probability mass in a region that encloses the two instances in every dimension. It deems two instances in a sparse region to be more similar than two instances of equal inter-point geometric distance in a dense region. Our empirical results in k-NN classification and content-based multimedia information retrieval tasks show that the proposed \(m_p\)-dissimilarity measure produces better task-specific performance than existing widely used general-purpose distance measures such as \(\ell _p\)-norm and cosine distance across a wide range of moderate- to high-dimensional data sets with continuous only, discrete only, and mixed attributes.     

Wafer-level vacuum package of two-dimensional micro-scanner

Microsystem Technologies - We present wafer-level vacuum package of two-dimensional (2-D) micro-scanner based on glass-silicon anodic bonding. To form the sacrificial gap for evacuating air in the...     

Fabrication and Characterization of Silicon (100) Membranes for a Multi-beam Superconducting Heterodyne Receiver

We fabricated silicon (100) membranes of 3 mm in diameter on the surface of silicon-on-insulator (SOI) substrates and investigated the characteristics of the membranes. The handle layer of one SOI substrate was etched using deep reactive ion etching process with the buried oxide (BOX) layer that remained together with the device layer. The BOX layer of the other SOI substrate was removed using C₄F₈-based plasma etching after the handle layer etching. The surfaces of both silicon (100) membranes were observed using the scanning white light interferometer system at room temperature. Both silicon (100) membranes have dome-like deformations. The silicon (100) membranes are effectively flattened by etching the BOX layer under the device layer. Both silicon (100) membranes were cooled from room temperature to 4 K by a Gifford–McMahon refrigerator. Wrinkles appeared on the surfaces of both silicon (100) membranes when the temperature dropped to about 200 K. However, the wrinkles disappeared below about 180 K. This phenomenon indicates the wrinkles at low temperature would depend on the properties of the silicon (100) of the device layers and independent of the properties of the BOX layers under the silicon (100) membranes.     

An approach to establishing a method for calculating inventory

In periodic review inventory systems, inventory is classified into cycle stock and safety stock. Cycle stock is defined as inventory that absorbs differences between supply and demand frequencies. It can be calculated without deficiency or excess because a method has been established for ensuring that the minimum on-hand inventory during a periodic review is zero. Safety stock is defined as inventory that absorbs various differences between supply and demand. Unlike for cycle stock, a method for calculating safety stock without deficiency or excess remains to be established. An approach is proposed to establishing a method for calculating inventory in which inventory is classified on the basis of the holding purpose and the calculation factors indicate solutions. This approach was applied to inventory held to absorb, on the basis of fluctuations in demand, the difference in terms of time and quantity between supply and demand. Stock held for this purpose is referred to as ‘fluctuation stock’. The objective is to establish a method for calculating fluctuation stock so that the minimum on-hand inventory during a periodic review is zero and to clarify the relationship between fluctuation stock and safety stock.     

Ellipsoidal load‐domain shakedown analysis with von Mises yield criterion: A robust optimization approach

The static formulation of elastic shakedown analysis, based upon Melan's lower‐bound theorem, can essentially be viewed as a robust optimization problem. This paper discusses an advantage that is enjoyed by taking this perspective. Specifically, we assume the von Mises yield criterion and an ellipsoidal load domain. In this setting, the shakedown analysis problem, which is viewed as robust second‐order cone programming, can be recast as semidefinite programming. Copyright © 2016 John Wiley & Sons, Ltd.     

Analog Circuit Applications Based on All‐2D Ambipolar ReSe2 Field‐Effect Transistors

Complementary circuits based on 2D materials show great promise for next‐generation electronics. An ambipolar all‐2D ReSe₂ field‐effect transistor (FET) with a hexagonal boron nitride gate dielectric is fabricated and its electronic characteristics are comprehensively studied by temperature dependence and noise measurements. Ambipolar transfer characteristics are achieved owing to the tunable Fermi level of the graphene contact and negligible and 30 meV Schottky barrier heights for the n‐ and p‐channel regimes, respectively. An inverter is also fabricated to demonstrate ambipolar ReSe₂ FET operation in a logic circuit. Furthermore, a p/n switchable unipolar FET is designed and shows potential for building complimentary circuits from a signal device. This work demonstrates the potential of all‐2D ReSe₂ FETs and makes available new approaches for designing next‐generation devices.     

Lattice Dynamics,Phonon Chirality,and Spin–Phonon Coupling in 2D Itinerant Ferromagnet Fe3GeTe2

Fe₃GeTe₂ has emerged as one of the most fascinating van der Waals crystals due to its 2D itinerant ferromagnetism, topological nodal lines, and Kondo lattice behavior. However, lattice dynamics, chirality of phonons, and spin–phonon coupling in this material, which set the foundation for these exotic phenomena, have remained unexplored. Here, the first experimental investigation of the phonons and mutual interactions between spin and lattice degrees of freedom in few‐layer Fe₃GeTe₂ is reported. The results elucidate three prominent Raman modes at room temperature: two A_1g(Γ) and one E_2g(Γ) phonons. The doubly degenerate E_2g(Γ) mode reverses the helicity of incident photons, indicating the pseudoangular momentum and chirality. Through analysis of temperature‐dependent phonon energies and lifetimes, which strongly diverge from the anharmonic model below Curie temperature, the spin–phonon coupling in Fe₃GeTe₂ is determined. Such interaction between lattice oscillations and spin significantly enhances the Raman susceptibility, allowing to observe two additional Raman modes at the cryogenic temperature range. In addition, laser radiation‐induced degradation of Fe₃GeTe₂ in ambient conditions and the corresponding Raman fingerprint is revealed. The results provide the first experimental analysis of phonons in this novel 2D itinerant ferromagnet and their applicability for further fundamental studies and application development.     

Scalability and homogeneity of slot die‐coated metal oxide semiconductor for TFTs

An indium oxide‐based precursor solution has been developed by spin coating method. In order to apply this material to mass production, material, process, and equipment optimizations for slot die coating have been implemented. Slot die coating is a cost‐effective and scalable process and already applied to photoresist materials in the display industry. The indium oxide‐based precursor solution has been coated on bare glasses and thin‐film transistor substrates by a mass production‐type slot die coater. Mobility of over 10 cm²/Vs is achieved for the first time for a large area at an annealing temperature of 350 °C. The homogeneity of the film will be presented.     

ALife approach to eco-evo-devo using evolution of virtual creatures

Evolutionary developmental biology, evo-devo, emerged to integrate evolution and development in the 1980s, in which evolution is conceptualized as heritable changes in development. Recently, the field is moving to a new synthesis: ecological evolutionary developmental biology, eco-evo-devo. We believe that artificial life (ALife) approach will provide new insights into the field, and also contribute to the field of robotics through the emergence perspective and the constructive methodology. This paper explores the potential of such an artificial life approach based on the evolution of virtual creatures by presenting our ongoing studies with three models: (1) metamorphosis model, (2) exaptation model and (3) Prey-predator model.     

Macroscopic fundamental diagram in simple model of urban traffic

Macroscopic fundamental diagram (MFD) is a reproducible unimodal relation between average vehicle density and average flow rate in urban traffic. Although this idea is tested by a few observations and simulations, its mechanism is not well understood. In order to understand it, a simple graph-based model of urban traffic is proposed. MFDs in our model system are investigated numerically for grid networks. It is found that MFDs in our system are discontinuous, which is inconsistent with the observation of real urban traffic.