Search

Approximate Nearest Neighbor Search (ANNS) plays a critical role in various disciplines spanning data mining and artificial intelligence, from information retrieval and computer vision to natural language processing and recommender systems. Data volumes have soared in recent years and the computational cost of an exhaustive exact nearest neighbor search is often prohibitive, necessitating the adoption of approximate techniques. The balanced performance and recall of graph-based approaches have more recently garnered significant attention in ANNS algorithms, however, only a few studies have explored harnessing the power of GPUs and multi-core processors despite the widespread use of massively parallel and general-purpose computing. To bridge this gap, we introduce a novel parallel computing hardware-based proximity graph and search algorithm. By leveraging the high-performance capabilities of modern hardware, our approach achieves remarkable efficiency gains. In particular, our method surpasses existing CPU and GPU-based methods in constructing the proximity graph, demonstrating higher throughput in both large- and small-batch searches while maintaining compatible accuracy. In graph construction time, our method, CAGRA, is 2.2~27x faster than HNSW, which is one of the CPU SOTA implementations. In large-batch query throughput in the 90% to 95% recall range, our method is 33~77x faster than HNSW, and is 3.8~8.8x faster than the SOTA implementations for GPU. For a single query, our method is 3.4~53x faster than HNSW at 95% recall., Comment: Accepted to ICDE 2024

In this paper, we propose cuSLINK, a novel and state-of-the-art reformulation of the SLINK algorithm on the GPU which requires only $O(Nk)$ space and uses a parameter $k$ to trade off space and time. We also propose a set of novel and reusable building blocks that compose cuSLINK. These building blocks include highly optimized computational patterns for $k$-NN graph construction, spanning trees, and dendrogram cluster extraction. We show how we used our primitives to implement cuSLINK end-to-end on the GPU, further enabling a wide range of real-world data mining and machine learning applications that were once intractable. In addition to being a primary computational bottleneck in the popular HDBSCAN algorithm, the impact of our end-to-end cuSLINK algorithm spans a large range of important applications, including cluster analysis in social and computer networks, natural language processing, and computer vision. Users can obtain cuSLINK at https://docs.rapids.ai/api/cuml/latest/api/#agglomerative-clustering, Comment: To appear in ECML PKDD 2023 by Springer Nature

The nEXO neutrinoless double beta decay experiment is designed to use a time projection chamber and 5000 kg of isotopically enriched liquid xenon to search for the decay in $^{136}$Xe. Progress in the detector design, paired with higher fidelity in its simulation and an advanced data analysis, based on the one used for the final results of EXO-200, produce a sensitivity prediction that exceeds the half-life of $10^{28}$ years. Specifically, improvements have been made in the understanding of production of scintillation photons and charge as well as of their transport and reconstruction in the detector. The more detailed knowledge of the detector construction has been paired with more assays for trace radioactivity in different materials. In particular, the use of custom electroformed copper is now incorporated in the design, leading to a substantial reduction in backgrounds from the intrinsic radioactivity of detector materials. Furthermore, a number of assumptions from previous sensitivity projections have gained further support from interim work validating the nEXO experiment concept. Together these improvements and updates suggest that the nEXO experiment will reach a half-life sensitivity of $1.35\times 10^{28}$ yr at 90% confidence level in 10 years of data taking, covering the parameter space associated with the inverted neutrino mass ordering, along with a significant portion of the parameter space for the normal ordering scenario, for almost all nuclear matrix elements. The effects of backgrounds deviating from the nominal values used for the projections are also illustrated, concluding that the nEXO design is robust against a number of imperfections of the model., Comment: 26 pages, 19 figures, version accepted by Journal of Phys. G

Nuclei import and export proteins, including cell cycle regulators. These import-export processes are modulated periodically by the cell cycle, for example due to the periodic assembly and breakdown of the nuclear envelope. As such, replicated DNA can be segregated between the two daughter cells and the proteins that were localized in the nucleus are free to diffuse throughout the cytoplasm. Here, we study a mathematical import-diffusion model to show how proteins, i.e. cell cycle regulators, could be redistributed in the cytoplasm by nuclei that periodically toggle between interphase and mitosis. We show that when the cell cycle period depends on the local concentration of regulators, the model exhibits mitotic waves. We discuss how the velocity and spatial origin of these mitotic waves depend on the different model parameters. This work is motivated by recent in vitro experiments reporting on mitotic waves in cycling cell-free extracts made with Xenopus laevis frog eggs, where multiple nuclei share the same cytoplasm. Such experiments have shown that nuclei act as pacemakers for the cell cycle and thus play an important role in collectively defining the spatial origin of mitotic waves., Comment: 16 pages, 7 figures

Silicon photomultipliers are regarded as a very promising technology for next-generation, cutting-edge detectors for low-background experiments in particle physics. This work presents systematic reflectivity studies of Silicon Photomultipliers (SiPM) and other samples in liquid xenon at vacuum ultraviolet (VUV) wavelengths. A dedicated setup at the University of M\"unster has been used that allows to acquire angle-resolved reflection measurements of various samples immersed in liquid xenon with 0.45{\deg} angular resolution. Four samples are investigated in this work: one Hamamatsu VUV4 SiPM, one FBK VUV-HD SiPM, one FBK wafer sample and one Large-Area Avalanche Photodiode (LA-APD) from EXO-200. The reflectivity is determined to be 25-36% at an angle of incidence of 20{\deg} for the four samples and increases to up to 65% at 70{\deg} for the LA-APD and the FBK samples. The Hamamatsu VUV4 SiPM shows a decline with increasing angle of incidence. The reflectivity results will be incorporated in upcoming light response simulations of the nEXO detector., Comment: 18 pages, 11 figures

High-performance primitives for mathematical operations on sparse vectors must deal with the challenges of skewed degree distributions and limits on memory consumption that are typically not issues in dense operations. We demonstrate that a sparse semiring primitive can be flexible enough to support a wide range of critical distance measures while maintaining performance and memory efficiency on the GPU. We further show that this primitive is a foundational component for enabling many neighborhood-based information retrieval and machine learning algorithms to accept sparse input. To our knowledge, this is the first work aiming to unify the computation of several critical distance measures on the GPU under a single flexible design paradigm and we hope that it provides a good baseline for future research in this area. Our implementation is fully open source and publicly available as part of the RAFT library of GPU-accelerated machine learning primitives (https://github.com/rapidsai/raft).

nEXO is a proposed tonne-scale neutrinoless double beta decay ($0\nu\beta\beta$) experiment using liquid ${}^{136}Xe$ (LXe) in a Time Projection Chamber (TPC) to read out ionization and scintillation signals. Between the field cage and the LXe vessel, a layer of LXe ("skin" LXe) is present, where no ionization signal is collected. Only scintillation photons are detected, owing to the lack of optical barrier around the field cage. In this work, we show that the light originating in the skin LXe region can be used to improve background discrimination by 5% over previous published estimates. This improvement comes from two elements. First, a fraction of the $\gamma$-ray background is removed by identifying light from interactions with an energy deposition in the skin LXe. Second, background from ${}^{222}Rn$ dissolved in the skin LXe can be efficiently rejected by tagging the $\alpha$ decay in the ${}^{214}Bi-{}^{214}Po$ chain in the skin LXe., Comment: 11 pages, 12 figures

The Uniform Manifold Approximation and Projection (UMAP) algorithm has become widely popular for its ease of use, quality of results, and support for exploratory, unsupervised, supervised, and semi-supervised learning. While many algorithms can be ported to a GPU in a simple and direct fashion, such efforts have resulted in inefficient and inaccurate versions of UMAP. We show a number of techniques that can be used to make a faster and more faithful GPU version of UMAP, and obtain speedups of up to 100x in practice. Many of these design choices/lessons are general purpose and may inform the conversion of other graph and manifold learning algorithms to use GPUs. Our implementation has been made publicly available as part of the open source RAPIDS cuML library (https://github.com/rapidsai/cuml).

Smarter applications are making better use of the insights gleaned from data, having an impact on every industry and research discipline. At the core of this revolution lies the tools and the methods that are driving it, from processing the massive piles of data generated each day to learning from and taking useful action. Deep neural networks, along with advancements in classical ML and scalable general-purpose GPU computing, have become critical components of artificial intelligence, enabling many of these astounding breakthroughs and lowering the barrier to adoption. Python continues to be the most preferred language for scientific computing, data science, and machine learning, boosting both performance and productivity by enabling the use of low-level libraries and clean high-level APIs. This survey offers insight into the field of machine learning with Python, taking a tour through important topics to identify some of the core hardware and software paradigms that have enabled it. We cover widely-used libraries and concepts, collected together for holistic comparison, with the goal of educating the reader and driving the field of Python machine learning forward., Comment: Preprint of a manuscript accepted for publication in "Machine Learning with Python," a special issue of Information (ISSN 2078-2489)

Characterization of the vacuum ultraviolet (VUV) reflectance of silicon photomultipliers (SiPMs) is important for large-scale SiPM-based photodetector systems. We report the angular dependence of the specular reflectance in a vacuum of SiPMs manufactured by Fondazionc Bruno Kessler (FBK) and Hamamatsu Photonics K.K. (HPK) over wavelengths ranging from 120 nm to 280 nm. Refractive index and extinction coefficient of the thin silicon-dioxide film deposited on the surface of the FBK SiPMs are derived from reflectance data of a FBK silicon wafer with the same deposited oxide film as SiPMs. The diffuse reflectance of SiPMs is also measured at 193 nm. We use the VUV spectral dependence of the optical constants to predict the reflectance of the FBK silicon wafer and FBK SiPMs in liquid xenon.

Measurements of electron drift properties in liquid and gaseous xenon are reported. The electrons are generated by the photoelectric effect in a semi-transparent gold photocathode driven in transmission mode with a pulsed ultraviolet laser. The charges drift and diffuse in a small chamber at various electric fields and a fixed drift distance of 2.0 cm. At an electric field of 0.5 kV/cm, the measured drift velocities and corresponding temperature coefficients respectively are $1.97 \pm 0.04$ mm/$\mu$s and $(-0.69\pm0.05)$\%/K for liquid xenon, and $1.42 \pm 0.03$ mm/$\mu$s and $(+0.11\pm0.01)$\%/K for gaseous xenon at 1.5 bar. In addition, we measure longitudinal diffusion coefficients of $25.7 \pm 4.6$ cm$^2$/s and $149 \pm 23$ cm$^2$/s, for liquid and gas, respectively. The quantum efficiency of the gold photocathode is studied at the photon energy of 4.73 eV in liquid and gaseous xenon, and vacuum. These charge transport properties and the behavior of photocathodes in a xenon environment are important in designing and calibrating future large scale noble liquid detectors.

Understanding reflective properties of materials and photodetection efficiency (PDE) of photodetectors is important for optimizing energy resolution and sensitivity of the next generation neutrinoless double beta decay, direct detection dark matter, and neutrino oscillation experiments that will use noble liquid gases, such as nEXO, DARWIN, DarkSide-20k, and DUNE. Little information is currently available about reflectivity and PDE in liquid noble gases, because such measurements are difficult to conduct in a cryogenic environment and at short enough wavelengths. Here we report a measurement of specular reflectivity and relative PDE of Hamamatsu VUV4 silicon photomultipliers (SiPMs) with 50 micrometer micro-cells conducted with xenon scintillation light (~175 nm) in liquid xenon. The specular reflectivity at 15 deg. incidence of three samples of VUV4 SiPMs is found to be 30.4+/-1.4%, 28.6+/-1.3%, and 28.0+/-1.3%, respectively. The PDE at normal incidence differs by +/-8% (standard deviation) among the three devices. The angular dependence of the reflectivity and PDE was also measured for one of the SiPMs. Both the reflectivity and PDE decrease as the angle of incidence increases. This is the first measurement of an angular dependence of PDE and reflectivity of a SiPM in liquid xenon., Comment: 21 pages, 15 figures, 6 tables. As accepted by JINST

nEXO is a proposed experiment to search for the neutrino-less double beta decay ($0\nu\beta\beta$) of $^{136}$Xe in a tonne-scale liquid xenon time projection chamber (TPC). The nEXO TPC will be equipped with charge collection tiles to form the anode. In this work, the charge reconstruction performance of this anode design is studied with a dedicated simulation package. A multi-variate method and a deep neural network are developed to distinguish simulated $0\nu\beta\beta$ signals from backgrounds arising from trace levels of natural radioactivity in the detector materials. These simulations indicate that the nEXO TPC with charge-collection tiles shows promising capability to discriminate the $0\nu\beta\beta$ signal from backgrounds. The estimated half-life sensitivity for $0\nu\beta\beta$ decay is improved by $\sim$20$~(32)\%$ with the multi-variate~(deep neural network) methods considered here, relative to the sensitivity estimated in the nEXO pre-conceptual design report.

In this paper we report on the characterization of the Hamamatsu VUV4 (S/N: S13370-6152) Vacuum Ultra-Violet (VUV) sensitive Silicon Photo-Multipliers (SiPMs) as part of the development of a solution for the detection of liquid xenon scintillation light for the nEXO experiment. Various SiPM features, such as: dark noise, gain, correlated avalanches, direct crosstalk and Photon Detection Efficiency (PDE) were measured in a dedicated setup at TRIUMF. SiPMs were characterized in the range $163 \text{ } \text{K} \leq \text{T}\leq 233 \text{ } \text{K}$. At an over voltage of $3.1\pm0.2$ V and at $\text{T}=163 \text{ }\text{K}$ we report a number of Correlated Avalanches (CAs) per pulse in the $1 \upmu\text{s}$ interval following the trigger pulse of $0.161\pm0.005$. At the same settings the Dark-Noise (DN) rate is $0.137\pm0.002 \text{ Hz/mm}^{2}$. Both the number of CAs and the DN rate are within nEXO specifications. The PDE of the Hamamatsu VUV4 was measured for two different devices at $\text{T}=233 \text{ }\text{K}$ for a mean wavelength of $189\pm7\text{ nm}$. At $3.6\pm0.2$ V and $3.5\pm0.2$ V of over voltage we report a PDE of $13.4\pm2.6\text{ }\%$ and $11\pm2\%$, corresponding to a saturation PDE of $14.8\pm2.8\text{ }\%$ and $12.2\pm2.3\%$, respectively. Both values are well below the $24\text{ }\%$ saturation PDE advertised by Hamamatsu. More generally, the second device tested at $3.5\pm0.2$ V of over voltage is below the nEXO PDE requirement. The first one instead yields a PDE that is marginally close to meeting the nEXO specifications. This suggests that with modest improvements the Hamamatsu VUV4 MPPCs could be considered as an alternative to the FBK-LF SiPMs for the final design of the nEXO detector.

We report on the performance of silicon photomultiplier (SiPM) light sensors operating in electric field strength up to 30 kV/cm and at a temperature of 149K, relative to their performance in the absence of an external electric field. The SiPM devices used in this study show stable gain, photon detection efficiency, and rates of correlated pulses, when exposed to external fields, within the estimated uncertainties. No observable physical damage to the bulk or surface of the devices was caused by the exposure., Comment: 16 pages, 12 figures, 2 tables and two conferences (INPC2016 and TIPP2017)

The search for neutrinoless double beta decay probes the fundamental properties of neutrinos, including whether or not the neutrino and antineutrino are distinct. Double beta detectors are large and expensive, so background reduction is essential for extracting the highest sensitivity. The identification, or 'tagging', of the $^{136}$Ba daughter atom from double beta decay of $^{136}$Xe provides a technique for eliminating backgrounds in the nEXO neutrinoless double beta decay experiment. The tagging scheme studied in this work utilizes a cryogenic probe to trap the barium atom in solid xenon, where the barium atom is tagged via fluorescence imaging in the solid xenon matrix. Here we demonstrate imaging and counting of individual atoms of barium in solid xenon by scanning a focused laser across a solid xenon matrix deposited on a sapphire window. When the laser sits on an individual atom, the fluorescence persists for $\sim$30~s before dropping abruptly to the background level, a clear confirmation of one-atom imaging. No barium fluorescence persists following evaporation of a barium deposit to a limit of $\leq$0.16\%. This is the first time that single atoms have been imaged in solid noble element. It establishes the basic principle of a barium tagging technique for nEXO.

Future tonne-scale liquefied noble gas detectors depend on efficient light detection in the VUV range. In the past years Silicon Photomultipliers (SiPMs) have emerged as a valid alternative to standard photomultiplier tubes or large area avalanche photodiodes. The next generation double beta decay experiment, nEXO, with a 5 tonne liquid xenon time projection chamber, will use SiPMs for detecting the $178\,\text{nm}$ xenon scintillation light, in order to achieve an energy resolution of $\sigma / Q_{\beta\beta} = 1\, \%$. This paper presents recent measurements of the VUV-HD generation SiPMs from Fondazione Bruno Kessler in two complementary setups. It includes measurements of the photon detection efficiency with gaseous xenon scintillation light in a vacuum setup and dark measurements in a dry nitrogen gas setup. We report improved photon detection efficiency at $175\,\text{nm}$ compared to previous generation devices, that would meet the criteria of nEXO. Furthermore, we present the projected nEXO detector light collection and energy resolution that could be achieved by using these SiPMs., Comment: 11 pages, 13 figures, 2 tables

The projected performance and detector configuration of nEXO are described in this pre-Conceptual Design Report (pCDR). nEXO is a tonne-scale neutrinoless double beta ($0\nu\beta\beta$) decay search in $^{136}$Xe, based on the ultra-low background liquid xenon technology validated by EXO-200. With $\simeq$ 5000 kg of xenon enriched to 90% in the isotope 136, nEXO has a projected half-life sensitivity of approximately $10^{28}$ years. This represents an improvement in sensitivity of about two orders of magnitude with respect to current results. Based on the experience gained from EXO-200 and the effectiveness of xenon purification techniques, we expect the background to be dominated by external sources of radiation. The sensitivity increase is, therefore, entirely derived from the increase of active mass in a monolithic and homogeneous detector, along with some technical advances perfected in the course of a dedicated R&D program. Hence the risk which is inherent to the construction of a large, ultra-low background detector is reduced, as the intrinsic radioactive contamination requirements are generally not beyond those demonstrated with the present generation $0\nu\beta\beta$ decay experiments. Indeed, most of the required materials have been already assayed or reasonable estimates of their properties are at hand. The details described herein represent the base design of the detector configuration as of early 2018. Where potential design improvements are possible, alternatives are discussed. This design for nEXO presents a compelling path towards a next generation search for $0\nu\beta\beta$, with a substantial possibility to discover physics beyond the Standard Model., Comment: 182 pages, minor revisions

Multi-strain competition on networks is observed in many contexts, including infectious disease ecology, information dissemination or behavioral adaptation to epidemics. Despite a substantial body of research has been developed considering static, time-aggregated networks, it remains a challenge to understand the transmission of concurrent strains when links of the network are created and destroyed over time. Here we analyze how network dynamics shapes the outcome of the competition between an initially endemic strain and an emerging one, when both strains follow a susceptible-infected-susceptible dynamics, and spread at time scales comparable with the network evolution one. Using time-resolved data of close-proximity interactions between patients admitted to a hospital and medical health care workers, we analyze the impact of temporal patterns and initial conditions on the dominance diagram and coexistence time. We find that strong variations in activity volume cause the probability that the emerging strain replaces the endemic one to be highly sensitive to the time of emergence. The temporal structure of the network shapes the dominance diagram, with significant variations in the replacement probability (for a given set of epidemiological parameters) observed from the empirical network and a randomized version of it. Our work contributes towards the description of the complex interplay between competing pathogens on temporal networks., Comment: 9 pages, 4 figures

The next-generation Enriched Xenon Observatory (nEXO) is a proposed experiment to search for neutrinoless double beta ($0\nu\beta\beta$) decay in $^{136}$Xe with a target half-life sensitivity of approximately $10^{28}$ years using $5\times10^3$ kg of isotopically enriched liquid-xenon in a time projection chamber. This improvement of two orders of magnitude in sensitivity over current limits is obtained by a significant increase of the $^{136}$Xe mass, the monolithic and homogeneous configuration of the active medium, and the multi-parameter measurements of the interactions enabled by the time projection chamber. The detector concept and anticipated performance are presented based upon demonstrated realizable background rates., Comment: v2 as published

A new design for the anode of a time projection chamber, consisting of a charge-detecting "tile", is investigated for use in large scale liquid xenon detectors. The tile is produced by depositing 60 orthogonal metal charge-collecting strips, 3~mm wide, on a 10~\si{\cm} $\times$ 10~\si{\cm} fused-silica wafer. These charge tiles may be employed by large detectors, such as the proposed tonne-scale nEXO experiment to search for neutrinoless double-beta decay. Modular by design, an array of tiles can cover a sizable area. The width of each strip is small compared to the size of the tile, so a Frisch grid is not required. A grid-less, tiled anode design is beneficial for an experiment such as nEXO, where a wire tensioning support structure and Frisch grid might contribute radioactive backgrounds and would have to be designed to accommodate cycling to cryogenic temperatures. The segmented anode also reduces some degeneracies in signal reconstruction that arise in large-area crossed-wire time projection chambers. A prototype tile was tested in a cell containing liquid xenon. Very good agreement is achieved between the measured ionization spectrum of a $^{207}$Bi source and simulations that include the microphysics of recombination in xenon and a detailed modeling of the electrostatic field of the detector. An energy resolution $\sigma/E$=5.5\% is observed at 570~\si{keV}, comparable to the best intrinsic ionization-only resolution reported in literature for liquid xenon at 936~V/\si{cm}., Comment: 18 pages, 13 figures, as published

We introduce and compare new compression approaches to obtain regularized solutions of large linear systems which are commonly encountered in large scale inverse problems. We first describe how to approximate matrix vector operations with a large matrix through a sparser matrix with fewer nonzero elements, by borrowing from ideas used in wavelet image compression. Next, we describe and compare approaches based on the use of the low rank SVD, which can result in further size reductions. We describe how to obtain the approximate low rank SVD of the original matrix using the sparser wavelet compressed matrix. Some analytical results concerning the various methods are presented and the results of the proposed techniques are illustrated using both synthetic data and a very large linear system from a seismic tomography application, where we obtain significant compression gains with our methods, while still resolving the main features of the solutions.

This document is one of four individual case study reports presenting the qualitative findings from a five-year investigation of the impact of accountability reform on students with disabilities in four states, eight districts, and twenty schools. This case study report presents qualitative data, collected over a two-year period--2001 and 2002--from two levels of the New York public education systems on the impact of accountability reform on students with disabilities. This report is divided into five sections. The first section contains information regarding the methodology used to collect and analyze the state and local data presented in this report. Section II is an overview of the assessment and accountability systems in New York, including changes resulting from NCLB. In Section III, the authors provide demographic information on each of the study sites. In Section IV, they present their research findings at the state and district levels, and in Section V they present a discussion of the major findings for each embedded case study and address cross-site themes. (Contains 4 tables.)

This paper is one of four individual case study reports presenting the qualitative findings from a five-year investigation of the impact of accountability reform on students with disabilities in four states, eight districts, and twenty schools. This case study report presents qualitative data collected over a four-year period--2001-2004--from three levels of the public education systems, and provides an analysis of the impact of accountability reform on students with disabilities in one of the participating four states. This report is divided into five sections. Section I contains information regarding the methodology used to collect and analyze the state and local data presented in this report. Section II is an overview of the assessment and accountability systems in Texas, including changes resulting from NCLB. In Section III, the authors provide demographic information on each of the study sites. In Section IV they present their research findings at the state, district, and school level, and in Section V they present a discussion of the major findings for each embedded case study and address cross site themes. (Contains 2 tables.)

This paper is one of four individual case study reports presenting the qualitative findings from a five-year investigation of the impact of accountability reform on students with disabilities in four states, eight districts, and twenty schools. This case study report presents qualitative data collected over a four-year period--2001-2004--from three levels of the public education systems, and provides an analysis of the impact of accountability reform on students with disabilities in one of the participating four states. This report is divided into five sections. Section I contains information regarding the methodology used to collect and analyze the state and local data presented in this report. Section II is an overview of the assessment and accountability systems in Maryland, including changes resulting from NCLB. In Section III the authors provide demographic information on each of the study sites. In Section IV they present their research findings at the state, district, and school level, and in Section V they present a discussion of the major findings for each embedded case study and address cross-site themes. (Contains 2 tables.)

This paper is one of four individual case study reports presenting the qualitative findings from a five-year investigation of the impact of accountability reform on students with disabilities in four states, eight districts, and twenty schools. This case study report presents qualitative data collected over a four-year period--2001-2004--from three levels of the public education systems. This report is divided into five sections. Section I contains information regarding the methodology used to collect and analyze the state and local data presented in this report. Section II is an overview of the assessment and accountability systems in California, including changes resulting from No Child Left Behind Act. In Section III the authors provide demographic information on each of the study sites. In Section IV they present their research findings at the state, district, and school level, and in Section V they present a discussion of the major findings for each embedded case study and address cross-site themes. (Contains 2 tables.)

We propose a class of spherical wavelet bases for the analysis of geophysical models and forthe tomographic inversion of global seismic data. Its multiresolution character allows for modeling with an effective spatial resolution that varies with position within the Earth. Our procedure is numerically efficient and can be implemented with parallel computing. We discuss two possible types of discrete wavelet transforms in the angular dimension of the cubed sphere. We discuss benefits and drawbacks of these constructions and apply them to analyze the information present in two published seismic wavespeed models of the mantle, for the statistics and power of wavelet coefficients across scales. The localization and sparsity properties of wavelet bases allow finding a sparse solution to inverse problems by iterative minimization of a combination of the $\ell_2$ norm of data fit and the $\ell_1$ norm on the wavelet coefficients. By validation with realistic synthetic experiments we illustrate the likely gains of our new approach in future inversions of finite-frequency seismic data and show its readiness for global seismic tomography., Comment: 43 pages, 11 figures, submitted to Geophysical Journal International

The effects of several nonlinear regularization techniques are discussed in the framework of 3D seismic tomography. Traditional, linear, $\ell_2$ penalties are compared to so-called sparsity promoting $\ell_1$ and $\ell_0$ penalties, and a total variation penalty. Which of these algorithms is judged optimal depends on the specific requirements of the scientific experiment. If the correct reproduction of model amplitudes is important, classical damping towards a smooth model using an $\ell_2$ norm works almost as well as minimizing the total variation but is much more efficient. If gradients (edges of anomalies) should be resolved with a minimum of distortion, we prefer $\ell_1$ damping of Daubechies-4 wavelet coefficients. It has the additional advantage of yielding a noiseless reconstruction, contrary to simple $\ell_2$ minimization (`Tikhonov regularization') which should be avoided. In some of our examples, the $\ell_0$ method produced notable artifacts. In addition we show how nonlinear $\ell_1$ methods for finding sparse models can be competitive in speed with the widely used $\ell_2$ methods, certainly under noisy conditions, so that there is no need to shun $\ell_1$ penalizations., Comment: 23 pages, 7 figures. Typographical error corrected in accelerated algorithms (14) and (20)

We propose the use of $\ell_1$ regularization in a wavelet basis for the solution of linearized seismic tomography problems $Am=d$, allowing for the possibility of sharp discontinuities superimposed on a smoothly varying background. An iterative method is used to find a sparse solution $m$ that contains no more fine-scale structure than is necessary to fit the data $d$ to within its assigned errors., Comment: 19 pages, 14 figures. Submitted to GJI July 2006. This preprint does not use GJI style files (which gives wrong received/accepted dates). Corrected typo

A study examined geography texts to evaluate their adequacy for meeting the diverse needs of students and to recommend modifications that will address specific deficits. Four geography textbooks were selected based on consultations with publishers, teachers, and school administrators to be representative of geography textbooks adopted in the United States. One chapter from each text that related to China, India, Philippines, Russia, and West Indies was selected and one lesson from each chapter was scrutinized. Evaluation forms were designed, definitional criteria were devised for evaluating geography instruction in each textbook, and four education doctoral students were trained to read and evaluate the texts using the form. Only one text included all the instructional elements examined. None of the instructional elements was present in all the lessons examined. It seems clear that these texts would pose a significant challenge for students who enter middle school with reading skills below grade level or for students who speak English as a second language. Although the textbooks are intended for middle school students, readability is around tenth-grade level. Only one text provided specific suggestions for accommodating diverse learners. Textbooks varied in the extent to which they prompted students to engage in complex thinking associated with geography, a domain in which many problems are ill-structured. Texts were dense with facts but contained few concepts or principles, and students were asked to reiterate or summarize information more often than they were asked to illustrate, predict, evaluate, or apply information. Textbooks analyzed provided little support for teaching complex thinking or employing effective teaching strategies. Contains 4 tables of data and 12 references. (BT)