Search

The analysis of APCs and Gold OA publishing suggests there may be a significant correlation, at the country-level, in APC expenses and the citation impact of resulting publications. High APC-values are then a disadvantage for low and middle-income countries, and they are even worse than what their nominal value reveals, as exchange rates based on purchase power parity indicators show how much more costly it can be to publish in US dollars. Thus, while OA publishing was intended to further equality, we now witness a higher chance of scholars being excluded from publishing.

SUMMARY Receiver functions have been used for decades to study the Earth’s major discontinuities by focusing on converted waves. Deconvolution, which is the mathematical backbone of the method, is assumed to remove the source time function and the far-field dependence on structure, making it a useful method to map the nearby Earth structure and its discontinuities. Ray theory, a plane incoming wavefield, and a sufficiently well-known near-receiver background velocity model are conventionally assumed to map the observations to locations in the subsurface. Many researchers are aware of the shortcoming of these assumptions and several remedies have been proposed for mitigating their consequences. Adjoint tomography with a quasi-exact forward operator is now within reach for most researchers, and we believe is the way forward in receiver function studies. A first step is to calculate adjoint sensitivity kernels for a given misfit function. Here, we derive the adjoint source for a receiver function waveform misfit. Using a spectral element forward code, we have calculated sensitivity kernels for P-to-S converted waves using several 2-D models representing an average crust with an underlying mantle. The kernels show profound differences between P- and S-wave speed sensitivity. The sensitivity to P-wave speed is wide-ranging and related to the scattered P-wavefield which interferes with that of the P-to-S converted wave. The S-wave speed sensitivity is more local and mostly associated to potential locations of P-to-S conversion, although more distant sensitivity is also observed. Notably, there is virtually no sensitivity to impedance. We further observe the well-known trade-off between depth of the discontinuity and wave speed, but find that considering a longer waveform that includes more surface reverberations reduces this trade-off significantly.

Summary In this abstract, we review the gradient-based Markov Chain Monte Carlo (MCMC) and demonstrate its applicability in inferring the uncertainty in seismic inversion. There are many flavours of gradient-based MCMC; here, we will only focus on the Unadjusted Langevin algorithm (ULA) and Metropolis-Adjusted Langevin algorithm (MALA). We propose an adaptive step-length based on the Lipschitz condition within ULA to automate the tuning of step-length and suppress the Metropolis-Hastings acceptance step in MALA. We consider the linear seismic travel-time tomography problem as a numerical example to demonstrate the applicability of both methods.

Many modern artists engage in semiotic research. They ask the same questions semioticians ask, even if they answer them in the form of artworks, rather than academic papers – questions such as: What kinds of meanings can be made with specific semiotic modes and media, 1 and how? How can semiotic modes and media be combined into multimodal wholes? How can the reach of semiotic modes and media be extended and how can new semiotic modes and media be developed? What are the semiotic affordances 2 of relevant technologies, whether technologies of the hand, recording technologies such as photography and tape recording, or the new synthesizing technologies?

An important challenge in hyperspectral imaging tasks is to cope with the large number of spectral bins. Common spectral data reduction methods do not take prior knowledge about the task into account. Consequently, sparsely occurring features that may be essential for the imaging task may not be preserved in the data reduction step. Convolutional neural network (CNN) approaches are capable of learning the specific features relevant to the particular imaging task, but applying them directly to the spectral input data is constrained by the computational efficiency. We propose a novel supervised deep learning approach for combining data reduction and image analysis in an end-to-end architecture. In our approach, the neural network component that performs the reduction is trained such that image features most relevant for the task are preserved in the reduction step. Results for two convolutional neural network architectures and two types of generated datasets show that the proposed Data Reduction CNN (DRCNN) approach can produce more accurate results than existing popular data reduction methods, and can be used in a wide range of problem settings. The integration of knowledge about the task allows for more image compression and higher accuracies compared to standard data reduction methods.

© 2019 Informa UK Limited, trading as Taylor & Francis Group. This paper opens by proposing a model for the holistic study of design software that emphasises user operations. It then examines the affordances of Explain Everything, a popular interactive whiteboard software, for animating concepts that play a crucial role in the Australian Year 7–10 physics curriculum. Building on earlier work on animation by Leão [(2012a). “Movement in Film Titles: An Analytical Approach.” Unpublished PhD, University of Technology, Sydney], the meaning-making resources of animation, in general, are presented as a stratified system (of choices), which is then deployed to explore the animation operations that Explain Everything facilitates. It is found that Explain Everything can animate some of the Year 7–10 physics concepts but has limitations with respect to others and is not able to realise all the options which the animation system generalises. In this way, the paper seeks to contribute to the growing body of social semiotic studies of multimodal semiotic technologies.

Ultrasonic imaging algorithms used in many clinical and industrial applications consist of three steps: A data preprocessing, an image formation and an image post-processing step. For efficiency, image formation often relies on an approximation of the underlying wave physics. A prominent example is the Delay-And-Sum (DAS) algorithm used in reflectivity-based ultrasonic imaging. Recently, deep neural networks (DNNs) are being used for the data pre-processing and the image postprocessing steps separately. In this work, we propose a novel deep learning architecture that integrates all three steps to enable endto- end training. We examine turning the DAS image formation method into a network layer that connects data pre-processing layers with image post-processing layers that perform segmentation. We demonstrate that this integrated approach clearly outperforms sequential approaches that are trained separately. While network training and evaluation is performed only on simulated data, we also showcase the potential of our approach on real data from a non-destructive testing scenario. Index Terms—deep learning, end-to-end training, Delay-And- Sum, fast ultrasonic imaging, approximate inversion.

In many ultrasonic imaging systems, data acquisition and image formation are performed on separate computing devices. Data transmission is becoming a bottleneck, thus, efficientdata compression is essential. Compression rates can be improved by considering the fact that many image formation methods rely on approximations of wave-matter interactions, and only use the corresponding part of the data. Tailored data compression could exploit this, but extracting the useful part of the data efficiently is not always trivial. In this work, we tackle this problem using deep neural networks, optimized to preserve the image quality of a particular image formation method. The Delay-And-Sum (DAS) algorithm is examined which is used in reflectivity-based ultrasonic imaging. We propose a novel encoder-decoder architecture with vector quantization and formulate image formation as a network layer for endto- end training. Experiments demonstrate that our proposed data compression tailored for a specific image formation method obtains significantly better results as opposed to compression agnostic to subsequent imaging. We maintain high image quality at much higher compression rates than the theoretical lossless compression rate derived from the rank of the linear imaging operator. This demonstrates the great potential of deep ultrasonic data compression tailored for a specific image formation method. Index Terms—deep learning, compression, Delay-And-Sum, fast ultrasonic imaging, end-to-end training

In this work we address regularization parameter estimation for ill-posed linear inverse problems with an penalty. Regularization parameter selection is of utmost importance for all of inverse problems and estimating it generally relies on the experience of the practitioner. For regularization with an penalty there exist a lot of parameter selection methods that exploit the fact that the solution and the residual can be written in explicit form. Parameter selection methods are functionals that depend on the regularization parameter where the minimizer is the desired regularization parameter that should lead to a good solution. Evaluation of these parameter selection methods still requires solving the inverse problem multiple times. Efficient evaluation of the parameter selection methods can be done through model order reduction. Two popular model order reduction techniques are Lanczos based methods (a Krylov subspace method) and the Randomized Singular Value Decomposition (RSVD). In this work we compare the two approaches. We derive error bounds for the parameter selection methods using the RSVD. We compare the performance of the Lanczos process versus the performance of RSVD for efficient parameter selection. The RSVD algorithm we use is based on the Adaptive Randomized Range Finder algorithm which allows for easy determination of the dimension of the reduced order model. Some parameter selection also require the evaluation of the trace of a large matrix. We compare the use of a randomized trace estimator versus the use of the Ritz values from the Lanczos process. The examples we use for our experiments are two model problems from geosciences.