Ultrasonic attenuation in layered superconductors /

We study the ultrasonic attenuation in layered superconductors using the Green's function formalism. General expressions are derived analytically and then calculated numerically by taking the nearest and next-nearest interactions in a disordered layered superconductor with random hoppings. Our results show huge anisotropics of ultrasonic attenuation in the superconductors and the strong dependence of ultrasonic attenuation on the temperature and the direction of polarization of the sound wave.

Structural analysis of interior layered deposits in north and south Coprates Chasma, Mars

Interior layered deposits within an embayment in the northern as well as near the southern wall of Coprates Chasma in the Valles Marineris, Mars are studied using HRSC, CTX, HiRISE and CRISM data. In the northern embayment, layered deposits outcrop in three separate locations (a western deposit, a central deposit and an eastern deposit). The central layered deposit in the north has a stratigraphic thickness of 2 km. The western layered deposit abuts against the chasma wall appearing to have a relatively un-eroded depositional surface. The eastern deposit is near a landslide scar which appears to have exposed basement layering showing downward displacement. This northern embayment is suggested to have been an ancestral basin. The triangular edged deposit near the southern wall of Coprates Chasma has an elongated mound protruding from the central edge and is suggested to be the outer limits of a fault block which is back rotated 6° south. The rotation may be the result of the Valles Marineris opening.

DEFORMATION OF INTERIOR LAYERED DEPOSITS (ILD) AND THEIR ASSOCIATED LANDFORMS IN WEST CANDOR CHASMA, MARS: IMPLICATIONS FOR THE TIMING OF ILD DEPOSITION

On Mars, interior layered deposits (ILD) provide evidence that water was once stable at the surface of the planet and present in large quantities. In West Candor Chasma, the ILD and their associated landforms record the depositional history of the chasma, and the deformation of those deposits provide insight into the stresses acting on them and the chasma as a whole. The post ILD structural history of West Candor is interpreted by analyzing the spatial relationships and orientation trends of structural features within the ILD. Therecording of stresses through brittle deformation of ILDs implies that the ILD had been lithified before the stress was imposed. Based on the prominent orientation trends of deformation features, the orientation of the stress regime acting upon the ILD appears to be linked to the regime that initially created the chasma-forming faults. An additional minor stress orientation was also revealed and may be related to large structures outside west Candor Chasma. The late depositional history of Ceti Mensa is herein investigated by examining the attributes and spatial relationship between unique corrugated, linear formations (CLF). The CLFs appear to be aeolian in origin but display clear indications of brittle deformation, indicating they have been Iithified. Evidence of lithification and the mineral composition of the surrounding material support the interpretation of circulating water in the area.

STRUCTURAL ANALYSIS, LAYER THICKNESS MEASUREMENTS AND MINERALOGICAL INVESTIGATION OF THE LARGE INTERIOR LAYERED DEPOSIT WITHIN GANGES CHASMA, VALLES MARINERIS, MARS

The interior layered deposit (ILD) in Ganges Chasma, Valles Marineris, is a 4.25 km high mound that extends approximately 110 km from west to east. The deposition, deformation, and erosion history of the Ganges ILD records aids in identifying the processes that formed and shaped the Chasma. To interpret structural and geomorphic processes acting on the ILD, multiple layer attitudes and layer thickness transects were conducted on the Ganges ILD. Mineralogical data was analyzed to determine correlations between materials and landforms. Layer thickness measurements indicate that the majority of layers are between 0.5 m and 4 m throughout the ILD. Three major benches dominate the Ganges ILD. Layer thicknesses increase at the ILD benches, suggesting that the benches are formed from the gradual thickening of layers. This indicates that the benches are depositional features draping over basement topography. Layer attitudes indicate overall shallow dips generally confined to a North-South direction that locally appear to follow bench topography. Layering is disrupted on a scale of 40 m to 150 m in 12 separate locations throughout the ILD. In all locations, underlying layering is disturbed by overlying folded layers in a trough-like geometry. These features are interpreted to have formed as submarine channels in a lacustrine setting, subsequently infilled by sediments. Subsequently, the channels were eroded to the present topography, resulting in the thin, curved layering observed. Data cannot conclusively support one ILD formation hypothesis, but does indicate that the Ganges ILD postdates Chasma formation. The presence of water altered minerals, consistently thin layering, and layer orientations provide strong evidence that the ILD formed in a lacustrine setting.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and deterministic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel metaheuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS metaheuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and determinis- tic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel meta–heuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS meta–heuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.