Using Discriminative Dimensionality Reduction to Visualize Classifiers

Neural Processing Letters - Albeit automated classifiers offer a standard tool in many application areas, there exists hardly a generic possibility to directly inspect their behavior, which goes...     

Flow-sensitive, context-sensitive, and object-sensitive information flow control based on program dependence graphs

Information flow control (IFC) checks whether a program can leak secret data to public ports, or whether critical computations can be influenced from outside. But many IFC analyses are imprecise, as they are flow-insensitive, context-insensitive, or object-insensitive; resulting in false alarms. We argue that IFC must better exploit modern program analysis technology, and present an approach based on program dependence graphs (PDG). PDGs have been developed over the last 20 years as a standard device to represent information flow in a program, and today can handle realistic programs. In particular, our dependence graph generator for full Java bytecode is used as the basis for an IFC implementation which is more precise and needs less annotations than traditional approaches. We explain PDGs for sequential and multi-threaded programs, and explain precision gains due to flow-, context-, and object-sensitivity. We then augment PDGs with a lattice of security levels and introduce the flow equations for IFC. We describe algorithms for flow computation in detail and prove their correctness. We then extend flow equations to handle declassification, and prove that our algorithm respects monotonicity of release. Finally, examples demonstrate that our implementation can check realistic sequential programs in full Java bytecode.     

Evolving trees for the retrieval of mass spectrometry-based bacteria fingerprints

In this paper, we investigate the application of Evolving Trees (ET) for the analysis of mass spectrometric data of bacteria. Evolving Trees are extensions of self-organizing maps (SOMs) developed for hierarchical classification systems. Therefore, they are well suited for taxonomic problems such as the identification of bacteria. Here, we focus on three topics, an appropriate pre-processing and encoding of the spectra, an adequate data model by means of a hierarchical Evolving Tree and an interpretable visualization. First, the high dimensionality of the data is reduced by a compact representation. Here, we employ sparse coding, specifically tailored for the processing of mass spectra. In the second step, the topographic information which is expected in the fingerprints is used for advanced tree evaluation and analysis. We adapted the original topographic product for SOMs for ET to achieve a judgment of topography. Additionally we transferred the concept of U-matrix for evaluation of the separability of SOMs to their analog in ET. We demonstrate these extensions for two mass spectrometric data sets of bacteria fingerprints and show their classification and evaluation capabilities in comparison to state of the art techniques.     

Reduced cross‐talk in shutter‐glass‐based stereoscopic LCD

Abstract— It is expected that 3‐D will be the next step in the enhanced viewing experience. At present, there are two competing 3‐D technologies for glasses‐based consumer TVs: active shutter glasses and passive polarized glasses. With the ongoing reduction in response time of liquid‐crystal displays (LCDs), this article will focus on shutter‐glass‐based stereoscopic LCDs. In this paper, the properties of such a display system is described and it is demonstrated that by adding a line‐scanning backlight, the cross‐talk can be reduced to less than 1.4%, allowing for excellent 3‐D portrayal. For images of extreme contrast, this is perceivable, but not judged annoying by a panel of expert viewers. Which characteristics of the display and shutter glasses that should be optimized to create an inexpensive, cross‐talk‐free, 3‐D LCD are discussed.     

An Internet-based negotiation server for e-commerce

This paper describes the design and implementation of a replicable, Internet-based negotiation server for conducting bargaining-type negotiations between enterprises involved in e-commerce and e-business. Enterprises can be buyers and sellers of products/services or participants of a complex supply chain engaged in purchasing, planning, and scheduling. Multiple copies of our server can be installed to complement the services of Web servers. Each enterprise can install or select a trusted negotiation server to represent his/her interests. Web-based GUI tools are used during the build-time registration process to specify the requirements, constraints, and rules that represent negotiation policies and strategies, preference scoring of different data conditions, and aggregation methods for deriving a global cost-benefit score for the item(s) under negotiation. The registration information is used by the negotiation servers to automatically conduct bargaining type negotiations on behalf of their clients. In this paper, we present the architecture of our implementation as well as a framework for automated negotiations, and describe a number of communication primitives which are used in the underlying negotiation protocol. A constraint satisfaction processor (CSP) is used to evaluate a negotiation proposal or counterproposal against the registered requirements and constraints of a client company. In case of a constraint violation, an event is posted to trigger the execution of negotiation strategic rules, which either automatically relax the violated constraint, ask for human intervention, invoke an application, or perform other remedial operations. An Event-Trigger-Rule (ETR) server is used to manage events, triggers, and rules. Negotiation strategic rules can be added or modified at run-time. A cost-benefit analysis component is used to perform quantitative analysis of alternatives. The use of negotiation servers to conduct automated negotiation has been demonstrated in the context of an integrated supply chain scenario. Received: 30 October 2000 / Accepted: 12 January 2001 Published online: 2 August 2001     

NOX removal in excess oxygen by plasma-enhanced selective catalytic reduction

In the off-gases of internal combustion engines running with oxygen excess, non-thermal plasmas (NTPs) have an oxidative potential, which results in an effective conversion of NO to NO₂. In combination with appropriate catalysts and ammonia (NH₃-SCR) or hydrocarbons (HC-SCR) as a reducing agent, this can be utilized to reduce nitric oxides (NO and NO₂) synergistically to molecular nitrogen.

The combination of SCR and cold plasma enhanced the overall reaction rate and allowed an effective removal of NO_X at low temperatures. Using NH₃ as a reducing agent, NO_X was converted to N₂ on zeolites or NH₃-SCR catalysts like V₂O₅–WO₃/TiO₂ at temperatures as low as 100–200 °C. Significant synergetic effects of plasma and catalyst treatment were observed both for NH₃ stored by ion exchange on the zeolite and for continuous NH₃ supply.

Certain modifications of Al₂O₃ and ZrO₂ have been found to be effective as catalysts in the plasma-assisted HC-SCR in oxygen excess. With an energy supply of about 30 eV/NO-molecule, 500 ppm NO was reduced by more than half at a temperature of 300 °C and a space velocity of 20 000 h⁻¹ at the catalyst. The synergistic combinations of NTP and both NH₃- and HC-SCR have been verified under real diesel engine exhaust conditions.     

A Strange Application of Kolmogorov Complexity

This paper deals with two similar inequalities: where K denotes simple Kolmogorov entropy (i.e., the very first version of Kolmogorov complexity having been introduced by Kolmogorov himself) and KP denotes prefix entropy (self-delimiting complexity by the terminology of Li and Vitanyi [1]). It turns out that from (1) the following well-known geometric fact can be inferred: where V is a set in three-dimensional space, S _xy , S _yz , S _xz are its three two-dimensional projections, and |W| is the volume (or the area) of W . Inequality (2), in its turn, is a corollary of the well-known Cauchy—Schwarz inequality. So the connection between geometry and Kolmogorov complexity works in both directions. Received April 20, 1993, and in final form December 6, 1993.     

COBRA-STAR, a five frame point-projection x-ray imaging system for 1 MA scale wire-array Z pinches

A new imaging system for 1 MA scale wire-array Z-pinch experiments that produces up to five high-resolution x-ray images per experimental pulse has been developed. Calibrated areal density measurements of the Z-pinch plasma can be obtained from each pulse. The system substitutes five molybdenum (Mo) X pinches for the normal copper return-current conductors to provide point sources of x-rays for point-projection radiography. Each backlighting X pinch consists of four Mo wires, the x-ray burst timing of which was controlled by varying the wire diameter (mass) from 10.2 to 30 microm in the five X pinches. Typical images have a 16x8 mm2 field of view at the wire array and a magnification of about 6.5:1 on the x-ray-sensitive film. Titanium (Ti) filters in front of the films transmit continuum radiation in the spectral range of 3-5 keV. Inclusion on the Ti of a step wedge having known thickness increments of the same material as the wires enables the calibrated areal density measurements to be made of the exploding wire plasmas. Here, we used tungsten (W) step wedges with step thicknesses ranging from 0.015 to 1.1 microm to obtain accurate (+/-10%) areal density measurements of W plasmas from the spatial profile of film exposure. When imaging arrays that produce intense radiation pulses, a plastic monofilament "quencher" is placed on axis to avoid film saturation. Images have subnanosecond temporal resolution and about 7 microm spatial resolution.