Small Number Of Configurations Research Articles

Selective Query Processing: A Risk-Sensitive Selection of Search Configurations

In information retrieval systems, search parameters are optimized to ensure high effectiveness based on a set of past searches, and these optimized parameters are then used as the search configuration for all subsequent queries. A better approach, however, would be to adapt the parameters to fit the query at hand. Selective query expansion is one such an approach, in which the system decides automatically whether or not to expand the query, resulting in two possible search configurations. This approach was extended recently to include many other parameters, leading to many possible search configurations where the system automatically selects the best configuration on a per-query basis. One problem with this approach is the system training, which requires evaluation of each training query with every possible configuration. In real-world systems, so many parameters and possible values must be evaluated that this approach is impractical, especially when the system must be updated frequently, as is the case for commercial search engines. In general, the more configurations, the greater the effectiveness when configuration selection is appropriate but also the greater the risk of decreasing effectiveness in the case of an inappropriate configuration selection. To determine the ideal configurations to be used for each query in real-world systems, we have developed a method in which a limited number of possible configurations are pre-selected, then used in a meta-search engine that decides the best search configuration for each query. We define a risk-sensitive approach for configuration pre-selection that considers the risk-reward tradeoff between the number of configurations kept and system effectiveness. We define two alternative risk functions to apply to different goals. For final configuration selection, the decision is based on query feature similarities. We compare two alternative risk functions on two query types (ad hoc and diversity) and compare these to more sophisticated machine learning based methods. We find that a relatively small number of configurations (20) selected by our risk-sensitive model is sufficient to obtain results close to the best achievable results for each query. Effectiveness is increased by about 15% according to the P@10 and nDCG@10 evaluation metrics when compared to traditional grid search using a single configuration and by about 20% when compared to learning to rank documents. Our risk-sensitive approach works for both diversity- and ad hoc oriented searches. Moreover, the similarity-based selection method outperforms the more sophisticated approaches. Thus, we demonstrate the feasibility of developing per-query information retrieval systems, which will guide future research in this direction.

Progressive Stereo Locking (PSL): A Residual Dipolar Coupling Based Force Field Method for Determining the Relative Configuration of Natural Products and Other Small Molecules.

Establishing the relative configuration of a bioactive natural product represents the most challenging part in determining its structure. Residual dipolar couplings (RDCs) are sensitive probes of the relative spatial orientation of internuclear vectors. We adapted a force field structure calculation methodology to allow free sampling of both R and S configurations of the stereocenters of interest. The algorithm uses a floating alignment tensor in a simulated annealing protocol to identify the conformations and configurations that best fit experimental RDC and distance restraints (from NOE and J-coupling data). A unique configuration (for rigid molecules) or a very small number of configurations (for less rigid molecules) of the structural models having the lowest chiral angle energies and reasonable magnitudes of the alignment tensor are provided as the best predictions of the unknown configuration. For highly flexible molecules, the progressive locking of their stereocenters into their statistically dominant R or S state dramatically reduces the number of possible relative configurations. The result is verified by checking that the same configuration is obtained by initiating the locking from different regions of the molecule. For all molecules tested having known configurations (with conformations ranging from mostly rigid to highly flexible), the method accurately determined the correct configuration.

Configuration interaction in symmetry-conserving covariant density functional theory

A new method to calculate spectroscopic properties of deformed nuclei is proposed: configuration interaction on top of projected density functional theory (CI-PDFT). The general concept of this approach is discussed in the framework of covariant density functional theory and its validity is illustrated in an application to the yrast band of the nucleus Cr-54. It is found that the experimentally observed excitation energies for the yrast band in Cr-54 can be well reproduced. In contrast to conventional shell-model calculations, there is no core and only a relatively small number of configurations is sufficient for a satisfying description. No new parameters are necessary, because the effective interaction is derived from an universal density functional given in the literature.

Monte Carlo configuration interaction with perturbation corrections for dissociation energies of first row diatomic molecules: C2, N2, O2, CO, and NO

Dissociation energies for the diatomic molecules C2, N2, O2, CO, and NO are estimated using the Monte Carlo configuration interaction (MCCI) and augmented by a second order perturbation theory correction. The calculations are performed using the correlation consistent polarized valence "triple zeta" atomic orbital basis and resulting dissociation energies are compared to coupled cluster calculations including up to triple excitations (CCSDT) and Full Configuration Interaction Quantum Monte Carlo (FCIQMC) estimates. It is found that the MCCI method readily describes the correct behavior for dissociation for the diatomics even when capturing only a relatively small fraction (∼80%) of the correlation energy. At this level only a small number of configurations, typically O(10(3)) from a FCI space of dimension O(10(14)), are required to describe dissociation. Including the perturbation correction to the MCCI estimates, the difference in dissociation energies with respect to CCSDT ranges between 1.2 and 3.1 kcal/mol, and the difference when comparing to FCIQMC estimates narrows to between 0.5 and 1.9 kcal/mol. Discussions on MCCI's ability to recover static and dynamic correlations and on the form of correlations in the electronic configuration space are presented.

Configuration selection in the MCSCF method. I. Application to the B1∑+ state of HF

Due to the relatively small number of configurations used in MC wavefunctions, proper selection of configurations is essential. Systematic selection methods, allowing for the anticipated changes in orbital characteristics, are developed for an excited state system, the B1∑+ state of HF. Configurations qualifying for inclusion into the MC wavefunction must have either a large coefficient in CI trial functions, or a large energy lowering relative to a few reference configurations. The MCSCF results for the B1∑+ state of HF compare well with CI results obtained here and obtained previously by other authors.

Efficient methods for finding transition states in chemical reactions: Comparison of improved dimer method and partitioned rational function optimization method

A combination of interpolation methods and local saddle-point search algorithms is probably the most efficient way of finding transition states in chemical reactions. Interpolation methods such as the growing-string method and the nudged-elastic band are able to find an approximation to the minimum-energy pathway and thereby provide a good initial guess for a transition state and imaginary mode connecting both reactant and product states. Since interpolation methods employ usually just a small number of configurations and converge slowly close to the minimum-energy pathway, local methods such as partitioned rational function optimization methods using either exact or approximate Hessians or minimum-mode-following methods such as the dimer or the Lanczos method have to be used to converge to the transition state. A modification to the original dimer method proposed by [Henkelman and Jonnson J. Chem. Phys. 111, 7010 (1999)] is presented, reducing the number of gradient calculations per cycle from six to four gradients or three gradients and one energy, and significantly improves the overall performance of the algorithm on quantum-chemical potential-energy surfaces, where forces are subject to numerical noise. A comparison is made between the dimer methods and the well-established partitioned rational function optimization methods for finding transition states after the use of interpolation methods. Results for 24 different small- to medium-sized chemical reactions covering a wide range of structural types demonstrate that the improved dimer method is an efficient alternative saddle-point search algorithm on medium-sized to large systems and is often even able to find transition states when partitioned rational function optimization methods fail to converge.

Free energy of solid solutions and phase diagrams via quasiharmonic lattice dynamics

We show how a configurational lattice dynamics technique, in which the free energy of a number of configurations is determined directly by means of a fully dynamic structural minimization, can be used to calculate thermodynamic properties of solid solutions and phase diagrams. No assumptions are made as to the nature of the solution and both configurational and vibrational entropy contributions are determined directly. Only a small number of configurations are required. We illustrate the method using MnO/MgO, for which our results support the recent experiments of Wood, Hackler, and Dobson [Contrib. Mineral. Petrol. 115, 438 (1994)] who, in contrast to previous workers, suggest the formation of a complete solid solution at temperatures only above 1100 K.

Measured and calculated effects of mutations in bacteriophage T4 lysozyme on interactions in solution

Understanding the molecular determinants of protein interactions in solution has fundamental implications for understanding protein solution thermodynamics and, hence, processes as diverse as separations performance and cellular self-organization. Our earlier theoretical calculations indicate that the protein-protein interactions are dominated by a small number of configurations in which highly complementary surface regions are apposed, rather than by the overall colloidal interactions. To examine this paradigm more explicitly, we investigated the effects of protein structural modifications on protein-protein interactions. Experimental measurements are presented of B(22)(') values of a set of mutants of Ser44 in bacteriophage T4 lysozyme. Effects are seen with both charged and uncharged substitutions. The results with the charged substitutions follow the expected trends, whereas those with the uncharged substitutions may be explained by the impact of the mutations on the local protein geometry, which directly affects the complementarity of protein interactions. These effects are also captured well by molecular calculations that account for the mutations. The interaction energetics between protein pairs could provide information on the propensity for adventitious interactions, which can have important implications for separations and for normal and pathological self-assembly. Thus, protein structural data implicit in genomic information, coupled with appropriate calculational and experimental tools, can ultimately provide insights into protein interactions in vivo and in bioprocessing.

Convergence study of multi-configuration time-dependent Hartree simulations: H2 scattering from LiF(001)

The multi-configurational time-dependent Hartree method is compared with close-coupling and close-coupling wave-packet calculations for a four-dimensional model of scattering of H2 from a LiF(001) surface. Accurate results are obtained for the rotation–diffraction scattering probabilities. An analysis of the accuracy of the simulations as a function of the number of configurations shows that more single-particle functions are needed for the scattering coordinate than for the others to obtain good results. Reasonable results are, however, already obtained with a small number of configurations.

Finite density results for Wilson fermions using the volume method

Nonzero chemical potential studies with Wilson fermions should avoid the proliferation of flavor-equivalent nucleon states encountered with staggered formulation of fermions. However, conventional wisdom has been that finite baryon density calculations with Wilson fermions will be prohibitively expensive. We demonstrate that the volume method applied to Wilson fermions gives surprisingly stable results on a small number of configurations. It is pointed out that this method may be applied to any local or nonlocal gauge invariant quantity. Some illustrative results for $<\overline{\psi}\psi>$ and $< J >$ at various values of $\mu$ in a quenched lattice simulation are given.

On Topology Preservation in 3D Thinning

Topology preservation is a major concern of parallel thinning algorithms for 2D and 3D binary images. To prove that a parallel thinning algorithm preserves topology, one must show that it preserves topology for all possible images. But it would be difficult to check all images, since there are too many possible images. Efficient sufficient conditions which can simplify such proofs for the 2D case were proposed by Ronse [ Discrete Appl. Math. 21, 1988, 69-79]. By Ronse′s results, a 2D parallel thinning algorithm can be proved to be topology preserving by checking a rather small number of configurations. This paper establishes sufficient conditions for 3D parallel thinning algorithms to preserve topology.

Quenched staggered spectrum at β = 6.0, 6.2 and 6.4

We give a preliminary report on the hadron spectrum on an ensemble of quenched lattices at β values of 6.0, 6.2 and 6.4, using staggered fermions and collaborators Rajan Gupta and Steve Sharpe. Because of the relatively small number of configurations we pay marginally more attention to the question of statistics than in previous analyses. We are unable to discredit quenched QCD.

Dissociation energetics of SiF systems of relevance to etching reactions

The bond energies of the 2Π and 4Σ− states of SiF, the 1Σ+ state of SiF+, the 3Σ− state of SiF−, the 1A1, 3B1, and 1B1 states of SiF2, the 2A1 state of SiF+2, and the ground states of SiH3–F, SiF3–F, SiH3–SiH3, and SiH3–SiF3 have been calculated using generalized valence bond and dissociation-consistent configuration interaction methods. Excellent bond energies are determined using only a relatively small number of configurations in the calculations. For example, the bond energy of the 2Π state of SiF is calculated to be 5.45 eV, in excellent agreement with the experimental value of 5.62 eV. For the inversion bend of SiF3 we predict that the transition state is a T-shaped configuration rather than one of D3h symmetry. Of note for the etching of solid silicon by reaction with F atoms is that unsaturated Si species react readily with F atoms to form strong bonds and that the bonding of a F atom to a lone pair of electrons on a Si atom is an activated process.

Theory of NMR relaxation in macromolecules: Restricted diffusion and jump models for multiple internal rotations in amino acid side chains

The theory required to extract detailed motional information from NMR relaxation times of nuclei in an amino acid side chain containing multiple internal rotation axes attached to a large macromolecule of at least cylindrical symmetry is developed. Emphasis is placed on the analysis of 13C-NMR of protonated carbons where dipolar relaxation is predominant. Extension to other relaxation processes is straightforward. The spectral density from which the relaxation times can be calculated is obtained for various models for the motion of the side chain. The existing theory which assumes that internal rotations are both independent and free is generalized to incorporate excluded volume effects in a heuristic way by restricting the amplitude of the internal rotations. It is found that small amplitude motions are ineffective in causing relaxation. Thus jump models involving a relatively small number of configurations are appropriate to describe the motion of the side chain. The advantage of jump models is twofold: (1) excluded volume effects can be handled simply by including only sterically allowed configurations, and (2) the unrealistic assumption of diffusional models that rotations about different bonds are independent can be removed by considering concerted motions in which the positions of only a small number of atoms are altered. An explicit expression for the spectral density, which is computationally practical when the number of configurations is less than 200, is obtained for a class of jump models in which the dynamics is described by a master equation. A detailed application of the general formalism is made to the case of a lysine side chain whose carbon–carbon backbone is constrained to lie on a tetrahedral lattice. Finally, numerical calculations are presented to illustrate some of the qualitative features of the different models, and the strategy that can be used to obtain motional information from experiment within the framework of these models is discussed.

Patterns of term structure in simple spectra.

view Abstract Citations (3) References Co-Reads Similar Papers Volume Content Graphics Metrics Export Citation NASA/ADS Patterns of term structure in simple spectra. Layzer, David Abstract Theoretically, a pure configuration of electrons outside closed subshells has a definite, calculable pattern of term structure. For example, the two intervals between the three terms of (closed sub- shells) (p1) have the theoretical ratio 1.5. The observed patterns, however, often vary widely from atom to atom. For example, the above-mentioned ratio for (Is ) ( p1) is about -5 in Bei and 8 in BIl. Discrepancies of this kind are generally, and in my opinion correctly, ascribed to the mixing of configurations. Two obstacles have stood in the way of a successful quantitative theoretical description of configuration mixing: the lack of reliable criteria for recognizing the important components in a configuration mixture; and the lack of workable mathematical machinery for calculating the degree of mixing of given configurations. In this paper I try to overcome both these obstacles. My first point is that the data can be grouped so as to exhibit useful regularities in the structure of configuration-sequences. A configuration-sequence is defined as a set of identical configurations in an iso-electronic sequence, e.g., (Is2) ( p2) in Bei, BIl, CIII If 1(Z) is the interval between two terms of a configuration in such a sequence, Z denoting the atomic number, then the interval-difference I (Z I) - 1(Z) becomes sensibly constant for moderately large Z. The set of all independent interval-differences for a given sequence characterizes the sequence as a whole. A simple theoretical argument now shows that the structure of this set is influenced only by a very special kind of configuration mixing, which I call stable mixing. Only a small number of configurations can participate in a stable mixture, and these may be readily enumerated in every instance. I have found a method of predicting the structure of the set of interval-differences characterizing a configuration-sequence, and have successfully applied it to the sequence (is ) ( p2). This method, moreover, may be used to predict patterns of term structure in individual atoms. The crux of the method lies in the construction of analytic one-electron wave functions of a very general form. The 15, 25, and 2p wave functions are as accurate as the corresponding numerical wave functions of Hartree and Fock. The variational wave functions are used as the basis for a perturbation treatment, which can be carried out to any desired degree of accuracy. The Observatory, University of Michigan, Ann Arbor, Mich. Publication: The Astronomical Journal Pub Date: April 1951 DOI: 10.1086/106496 Bibcode: 1951AJ.....56...43L full text sources ADS |