Numerical experience with the truncated Newton method for unconstrained optimization

The truncated Newton algorithm was devised by Dembo and Steihaug (Ref. 1) for solving large sparse unconstrained optimization problems. When far from a minimum, an accurate solution to the Newton equations may not be justified. Dembo's method solves these equations by the conjugate direction method, but truncates the iteration when a required degree of accuracy has been obtained. We present favorable numerical results obtained with the algorithm and compare them with existing codes for large-scale optimization.     

Induction of decision trees via evolutionary programming

Decision trees have been used extensively in cheminformatics for modeling various biochemical endpoints including receptor-ligand binding, ADME properties, environmental impact, and toxicity. The traditional approach to inducing decision trees based upon a given training set of data involves recursive partitioning which selects partitioning variables and their values in a greedy manner to optimize a given measure of purity. This methodology has numerous benefits including classifier interpretability and the capability of modeling nonlinear relationships. The greedy nature of induction, however, may fail to elucidate underlying relationships between the data and endpoints. Using evolutionary programming, decision trees are induced which are significantly more accurate than trees induced by recursive partitioning. Furthermore, when assessed on previously unseen data in a 10-fold cross-validated manner, evolutionary programming induced trees exhibit a significantly higher accuracy on previously unseen data. This methodology is compared to single-tree and multiple-tree recursive partitioning in two domains (aerobic biodegradability and hepatotoxicity) and shown to produce less complex classifiers with average increases in predictive accuracy of 5-10% over the traditional method.     

Dilepton rapidity distribution in the Drell-Yan process at next-to-next-to-leading order in QCD

We compute the rapidity distribution of the virtual photon produced in the Drell-Yan process through next-to-next-to-leading order in perturbative QCD. We introduce a powerful new method for calculating differential distributions in hard scattering processes. This method is based upon a generalization of the optical theorem; it allows the integration-by-parts technology developed for multiloop diagrams to be applied to noninclusive phase-space integrals, and permits a high degree of automation. We apply our results to the analysis of fixed-target experiments.     

First Cr(III)-SNS complexes and their use as highly efficient catalysts for the trimerization of ethylene to 1-hexene

Cr(III) complexes of tridentate SNS ligands have been prepared and evaluated as catalysts for ethylene trimerization, with several giving very high activity and excellent selectivity toward 1-hexene when activated with methylaluminoxane. The new complexes illustrate the potential of sulfur-based ligands on early transition metals for catalysis.     

Solution-phase synthesis of an aminomethyl-substituted biaryl library via sequential amine N-alkylation and Suzuki cross-coupling

Described herein is the semiautomated preparation of a 20-member aminomethyl-substituted biaryl library. The two-step solution-phase synthesis was achieved via sequential N-alkylation of various amines with either 3- or 4-bromobenzyl bromide and Suzuki cross-coupling of the resultant aryl bromides with various boronic acids.