Efficient tree-traversals: reconciling parallelism and dense data representations

Recent work showed that compiling functional programs to use dense, serialized memory representations for recursive algebraic datatypes can yield significant constant-factor speedups for sequential programs. But serializing data in a maximally dense format consequently serializes the processing of that data, yielding a tension between density and parallelism. This paper shows that a disciplined, practical compromise is possible. We present Parallel Gibbon, a compiler that obtains the benefits of dense data formats and parallelism. We formalize the semantics of the parallel location calculus underpinning this novel implementation strategy, and show that it is type-safe. Parallel Gibbon exceeds the parallel performance of existing compilers for purely functional programs that use recursive algebraic datatypes, including, notably, abstract-syntax-tree traversals as in compilers.

Modeling asteroid shapes using BNAO Rozhen photometric data in combination with sparse data

<p>Our work aims to demonstrate how the use of our dense lightcurves in combination with sparse data from diverse sources will affect the results for obtaining the sidereal period, shape models, and ecliptic pole solution for a chosen asteroid.</p> <p>Photometric observations of minor planets are traditional at the Bulgarian National Astronomical observatory (BNAO) Rozhen. They started with photoelectric observations in 1991, and later have been continued as CCD photometric observations on all three telescopes: 2m Ritchey-Chretién-Coudé, 50cm/70cm, and 60cm Cassegrain. We hope that the new 1.5 m robotic telescope planned to be operational next year will be also partly devoted to the study of minor planets.</p> <p>Our target, 339 Dorothea, is a main-belt asteroid, a large member of the Eos dynamical family. For the last 8 years, between 2013 and 2021, the asteroid 339 Dorothea was observed at BNAO Rozhen during six apparitions and several dense lightcurve were obtained. We used these dense photometric data in lightcurve inversion method and reconstruct the model of the asteroid, determining its sidereal period, shape, and pole orientation. Afterward, using sparse data from the AstDys database with an accuracy of 0.01 mag in combination with the obtained dense data, new trials for calculating and improving the physical characteristics of the asteroid 339 Dorothea were made.</p> <p>Unlike very low photometric accuracy in ground-based sparse photometry, space missions have provided astronomers with sparse photometry with extremely high accuracy, for example, the ESA GAIA mission. The NEOWISE mission has observations only for a limited number of asteroids. Fortunately, we were able to find some sparse data for our target and use this accurate photometry in combination with our dense lightcurves for the reconstruction of the asteroid spin state and shape model.</p> <p>Due to bad weather conditions and limited allocation of observing time at the BNAO Rozhen dedicated to our project, we have at our disposal full and partial dense lightcurves obtained for several more asteroids in few different apparitions. Combining these dense data with ground-based or space mission sparse data will contribute to enlarging the database of asteroids with known physical characteristics. Enriching the number of asteroids with known physical parameters would provide more data for future statistical analysis and could help in answering the questions for the evolution of our Solar System. </p>

Deep-learning-based seismic data interpolation: A preliminary result

Seismic data interpolation is a longstanding issue. Most current methods are only suitable for randomly missing cases. To deal with regularly missing cases, an antialiasing strategy should be included. However, seismic survey design using a random distribution of shots and receivers is always operationally challenging and impractical. We have used deep-learning-based approaches for seismic data antialiasing interpolation, which could extract deeper features of the training data in a nonlinear way by self-learning. It can also avoid linear events, sparsity, and low-rank assumptions of the traditional interpolation methods. Based on convolutional neural networks, eight-layers residual learning networks (ResNets) with a better back-propagation property for deep layers is designed for interpolation. Detailed training analysis is also performed. A set of simulated data is used to train the designed ResNets. The performance is assessed with several synthetic and field data. Numerical examples indicate that the trained ResNets can help to reconstruct regularly missing traces with high accuracy. The interpolated results in the time-space domain and the frequency-wavenumber ([Formula: see text]-[Formula: see text]) domain demonstrate the validity of the trained ResNets. Even though the accuracy decreases with the increase of the feature difference between the test and training data, the proposed method can still provide reasonable interpolation results. Finally, the trained ResNets is used to reconstruct dense data with halved trace intervals for synthetic and field data. The reconstructed dense data are more continuous along the spatial direction, and the spatial aliasing effects disappear in the [Formula: see text]-[Formula: see text] domain. The reconstructed dense data have the potential to improve the accuracy of subsequent seismic data processing and inversion.