Register Allocation in Structured Programs

In this article we look at the register allocation problem. In the literature this problem is frequently reduced to the general graph coloring problem and the solutions to the problem are obtained from graph coloring heuristics. Hence, no algorithm with a good performance guarantee is known. Here we show that when attention is restricted tostructured programswhich we define to be programs whose control-flow graphs are series-parallel, there is an efficient algorithm that produces a solution which is within a factor of 2 of the optimal solution. We note that even with the previous restriction the resulting coloring problem is NP-complete.We also consider how to delete a minimum number of edges from arbitrary control-flow graphs to make them series-parallel and to apply our algorithm. We show that this problem is Max SNP hard. However, we define the notion of anapproximate articulation pointand we give efficient algorithms to find approximate articulation points. We present a heuristic for the edge deletion problem based on this notion which seems to work well when the given graph is close to series-parallel.     

On the Interplay among Entropy, Variable Metrics and Potential Functions in Interior-Point Algorithms

We are motivated by the problem of constructing aprimal-dual barrier function whose Hessian induces the (theoreticallyand practically) popular symmetric primal and dual scalings forlinear programming problems. Although this goal is impossible toattain, we show that the primal-dual entropy function may provide asatisfactory alternative. We study primal-dual interior-pointalgorithms whose search directions are obtained from a potentialfunction based on this primal-dual entropy barrier. We providepolynomial iteration bounds for these interior-point algorithms. Thenwe illustrate the connections between the barrier function and areparametrization of the central path equations. Finally, we considerthe possible effects of more general reparametrizations oninfeasible-interior-point algorithms.     

Computers with Closed Timelike Curves Can Solve Hard Problems Efficiently

A computer which has access to a closed timelike curve, and can thereby send the results of calculations into its own past, can exploit this to solve difficult computational problems efficiently. I give a specific demonstration of this for the problem of factoring large numbers and argue that a similar approach can solve NP-complete and PSPACE-complete problems. I discuss the potential impact of quantum effects on this result.     

Toward low‐band gap dithienophosphole copolymers for an application in organic solar cells

Two functionalized dithieno[3,2‐b:2′,3′‐d]phospholes with solubilizing groups have been synthesized that allow for the generation of a series of π‐conjugated AB‐ and ABC‐copolymers. The polymers obtained show notable optoelectronic properties with red‐shifted absorption and emission in the orange to red section of the optical solar spectrum. Although combination of dithienophosphole units with fluorene building blocks gives access to processable polymers with band gaps between 2.2 and 2.3 eV in solution and 2.0 eV in the solid state, an ABC copolymer based on dithienophosphole, fluorene, and bis(thienyl)benzothiadiazole units was found to not only exhibit a suitable band gap for solar cell applications (solution: 2.0 eV; solid state: 1.7 eV) but also showed good solubility as well as good electron transfer properties in the presence of fullerene (C₆₀). © 2008 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 46: 8179–8190, 2008     

Luminescence of charge transfer sensitizers anchored to metal oxide nanoparticles

The photoluminescence (PL) properties of inorganic charge transfer sensitizers anchored to nanometer sized metal oxide particles are presented. The charge transfer sensitizers are inorganic coordination compounds such as ruthenium tribipyridine, Ru(bpy)²⁺₃, which have long lived metal-to-ligand charge transfer (MLCT) excited states. The metal oxides are insulators or semiconductor materials in the form of powders, colloidal solutions, and porous nanocrystalline films. Time resolved PL decays from this and related sensitizers anchored to metal oxide surfaces are highly non-exponential. The MLCT excited states are quenched on semiconducting metal oxide particles by an apparent electron transfer mechanism. With some assumptions electron transfer rates from the MLCT excited states to the nanostructured surface are calculated. The PL properties of sensitizers bound to porous nanocrystalline TiO₂ films can be controlled electrochemically.     

Back and forth error compensation and correction methods for semi-lagrangian schemes with application to level set interface computations

Semi-Lagrangian schemes have been explored by several authors recently for transport problems, in particular for moving interfaces using the level set method. We incorporate the backward error compensation method developed in our paper from 2003 into semi-Lagrangian schemes with almost the same simplicity and three times the complexity of a first order semi-Lagrangian scheme but with improved order of accuracy. Stability and accuracy results are proved for a constant coefficient linear hyperbolic equation. We apply this technique to the level set method for interface computation.

     

Morera theorems via microlocal analysis

We prove Morera theorems for curves in the plane using microlocal analysis. The key is that microlocal smoothness of functions is reflected by smoothness of their Morera integrals on curvestheir Radon transforms. Parallel support theorems for the associated Radon transforms follow from our arguments by a simple correspondence.     

Automated filter paper assay for determination of cellulase activity

Recent developments in molecular breeding and directed evolution have promised great developments in industrial enzymes as demonstrated by exponential improvements in β-lactamase and green fluorescent protein (GFP). Detection of and screening for improved enzymes are relatively easy if the target enzyme is expressible in a suitable high-throughput screening host and a clearly defined and usable screen or selection is available, as with GFP and β-lactamase. Fungal cellulases, however, are difficult to measure and have limited expressibility in heterologous hosts. Furthermore, traditional cellulase assays are tedious and time-consuming. Multiple enzyme components, an insoluble substrate, and generally slow reaction rates have plagued cellulase researchers interested in creating cellulase mixtures with increased activities and/or enhanced biochemical properties. Although the International Union of Pure and Applied Chemists standard measure of cellulase activity, the filter paper assay (FPA), can be reproduced in most laboratories with some effort, this method has long been recognized for its complexity and susceptibility to operator error. Our current automated FPA method is based on a Cyberlabs C400 robotics deck equipped with customized incubation, reagent storage, and plate-reading capabilities that allow rapid evaluation of cellulases acting on cellulose and has a maximum throughput of 84 enzyme samples per day when performing the automated FPA.