Climate simulations are essential in guiding our understanding of climate change and responding to its effects. However, it is computationally expensive to resolve complex climate processes at high spatial resolution. As one way to speed up climate simulations, neural networks have been used to downscale climate variables from fast-running low-resolution simulations, but high-resolution training data are often unobtainable or scarce, greatly limiting accuracy. In this work, we propose a downscaling method based on the Fourier neural operator. It is trained using a low upsampling factor and then can zero-shot (without additional training) downscale its input to arbitrary unseen high resolution. Evaluated both on ERA5 climate model data and on the Navier-Stokes equation solution data, our downscaling model significantly outperforms state-of-the-art convolutional and generative adversarial downscaling models, both in standard single-resolution downscaling and in zero-shot generalization to higher upsampling factors. Furthermore, we show that our method also outperforms state-of-the-art data-driven partial differential equation solvers on Navier-Stokes equations. Overall, our work bridges the gap between simulation of a physical process and interpolation of low-resolution output, showing that it is possible to combine both approaches and significantly improve upon each other.

The availability of reliable, high-resolution climate and weather data is important to inform long-term decisions on climate adaptation and mitigation and to guide rapid responses to extreme events. Forecasting models are limited by computational costs and, therefore, often generate coarse-resolution predictions. Statistical downscaling, including super-resolution methods from deep learning, can provide an efficient method of upsampling low-resolution data. However, despite achieving visually compelling results in some cases, such models frequently violate conservation laws when predicting physical variables. In order to conserve physical quantities, here we introduce methods that guarantee statistical constraints are satisfied by a deep learning downscaling model, while also improving their performance according to traditional metrics. We compare different constraining approaches and demonstrate their applicability across different neural architectures as well as a variety of climate and weather data sets. Besides enabling faster and more accurate climate predictions through downscaling, we also show that our novel methodologies can improve super-resolution for satellite data and natural images data sets.

Seismic interpretation is a complex procedure that depends on many and interdependent data analyses. One of the essential steps in this process is picking horizons in seismic images, which is time-consuming and prone to errors when performed manually. In this context, having a reliable horizon picking tool is fundamental for accurate seismic interpretation. Although several methods for horizon picking have been proposed in the literature and many tools made available in the industry, most require numerous iterations and manual corrections for delivering satisfactory results. In this article, we present a three-step approach that improves the robustness of horizon picking using state-of-art semantic segmentation neural networks followed by geometry-based processing of horizon points to identify and remove outliers. For the well-known Netherlands F3 Block data set, this approach allows accurate results from a few training annotated inlines and delivers geometry-consistent horizons through all the cubes. We further present results for an additional set comprising four cubes with different seismic structures to provide some robustness evidence of our approach. The results reported in this study indicate that the proposed methodology can provide a good trade-off between accuracy and time spent on manually picking horizons.

Deep neural networks have gained attention in the last decade as significant progress has been made in a variety of tasks thanks to these new architectures. Most of the time, hand-designed networks are responsible for this incredible success. However, this engineering process demands considerable time and expert knowledge, which leads to an increasing interest in automating the design of deep architectures. Several new algorithms have been proposed to address the neural architecture search problem, but many of them require significant computational resources. Quantum-inspired evolutionary algorithms (QIEA) have their roots on quantum computing principles and present promising results in respect to faster convergence. In this work, we propose Q-NAS (Quantum-inspired Neural Architecture Search): a quantum-inspired algorithm to search for deep neural architectures by assembling substructures and optimizing some numerical hyperparameters. We present the first results applying Q-NAS on the CIFAR-10 dataset using only 20 K80 GPUs for about 50 hours. The obtained networks are relatively small (less than 20 layers) compared to other state-of-the-art models and achieve promising accuracies with considerably less computational cost than other NAS algorithms.

One of the most critical activities for the oil and gas industry is the discovery of new possibles reserves. Geoscientists must rely on indirect measures of the subsurface to scrutinize huge areas looking for leads of hydrocarbon reservoirs. Usually, to study the Earth's crust, geoscientists examine seismic images. Although deep learning has become popular in the last decade, only a few published results have demonstrated the application of such techniques to seismic images. In this paper, we present deep neural models specifically for the task of seismic facies analysis, using state-of-the-art concepts and tools to train and classify seismic facies efficiently. Our results show that we can train a neural network in 4 minutes using less than 5% of the dataset, and yet obtain 88% of accuracy. Moreover, we can reach up to 97% of accuracy in 30 minutes using 60% of the dataset.