SAR-UNet: Small Attention Residual UNet for Explainable Nowcasting Tasks

The accuracy and explainability of data-driven nowcasting models are of great importance in many socio-economic sectors reliant on weather-dependent decision making. This paper proposes a novel architecture called Small Attention Residual UNet (SAR-UNet) for precipitation and cloud cover nowcasting. Here, SmaAt-UNet is used as a core model and is further equipped with residual connections, parallel to the depthwise separable convolutions. The proposed SAR-UNet model is evaluated on two datasets, i.e., Dutch precipitation maps ranging from 2016 to 2019 and French cloud cover binary images from 2017 to 2018. The obtained results show that SAR-UNet outperforms other examined models in precipitation nowcasting from 30 to 180 minutes in the future as well as cloud cover nowcasting in the next 90 minutes. Furthermore, we provide additional insights on the nowcasts made by our proposed model using Grad-CAM, a visual explanation technique, which is employed on different levels of the encoder and decoder paths of the SAR-UNet model and produces heatmaps highlighting the critical regions in the input image as well as intermediate representations to the precipitation. The heatmaps generated by Grad-CAM reveal the interactions between the residual connections and the depthwise separable convolutions inside of the multiple depthwise separable blocks placed throughout the network architecture.

WF-UNet: Weather Fusion UNet for Precipitation Nowcasting

Designing early warning systems for harsh weather and its effects, such as urban flooding or landslides, requires accurate short-term forecasts (nowcasts) of precipitation. Nowcasting is a significant task with several environmental applications, such as agricultural management or increasing flight safety. In this study, we investigate the use of a UNet core-model and its extension for precipitation nowcasting in western Europe for up to 3 hours ahead. In particular, we propose the Weather Fusion UNet (WF-UNet) model, which utilizes the Core 3D-UNet model and integrates precipitation and wind speed variables as input in the learning process and analyze its influences on the precipitation target task. We have collected six years of precipitation and wind radar images from Jan 2016 to Dec 2021 of 14 European countries, with 1-hour temporal resolution and 31 square km spatial resolution based on the ERA5 dataset, provided by Copernicus, the European Union's Earth observation programme. We compare the proposed WF-UNet model to persistence model as well as other UNet based architectures that are trained only using precipitation radar input data. The obtained results show that WF-UNet outperforms the other examined best-performing architectures by 22%, 8% and 6% lower MSE at a horizon of 1, 2 and 3 hours respectively.

MSCDA: Multi-level Semantic-guided Contrast Improves Unsupervised Domain Adaptation for Breast MRI Segmentation in Small Datasets

Deep learning (DL) applied to breast tissue segmentation in magnetic resonance imaging (MRI) has received increased attention in the last decade, however, the domain shift which arises from different vendors, acquisition protocols, and biological heterogeneity, remains an important but challenging obstacle on the path towards clinical implementation. Recently, unsupervised domain adaptation (UDA) methods have attempted to mitigate this problem by incorporating self-training with contrastive learning. To better exploit the underlying semantic information of the image at different levels, we propose a Multi-level Semantic-guided Contrastive Domain Adaptation (MSCDA) framework to align the feature representation between domains. In particular, we extend the contrastive loss by incorporating pixel-to-pixel, pixel-to-centroid, and centroid-to-centroid contrasts to integrate semantic information of images. We utilize a category-wise cross-domain sampling strategy to sample anchors from target images and build a hybrid memory bank to store samples from source images. Two breast MRI datasets were retrospectively collected: The source dataset contains non-contrast MRI examinations from 11 healthy volunteers and the target dataset contains contrast-enhanced MRI examinations of 134 invasive breast cancer patients. We set up experiments from source T2W image to target dynamic contrast-enhanced (DCE)-T1W image (T2W-to-T1W) and from source T1W image to target T2W image (T1W-to-T2W). The proposed method achieved Dice similarity coefficient (DSC) of 89.2\% and 84.0\% in T2W-to-T1W and T1W-to-T2W, respectively, outperforming state-of-the-art methods. Notably, good performance is still achieved with a smaller source dataset, proving that our framework is label-efficient.

BAST: Binaural Audio Spectrogram Transformer for Binaural Sound Localization

Accurate sound localization in a reverberation environment is essential for human auditory perception. Recently, Convolutional Neural Networks (CNNs) have been utilized to model the binaural human auditory pathway. However, CNN shows barriers in capturing the global acoustic features. To address this issue, we propose a novel end-to-end Binaural Audio Spectrogram Transformer (BAST) model to predict the sound azimuth in both anechoic and reverberation environments. Two modes of implementation, i.e. BAST-SP and BAST-NSP corresponding to BAST model with shared and non-shared parameters respectively, are explored. Our model with subtraction interaural integration and hybrid loss achieves an angular distance of 1.29 degrees and a Mean Square Error of 1e-3 at all azimuths, significantly surpassing CNN based model. The exploratory analysis of the BAST's performance on the left-right hemifields and anechoic and reverberation environments shows its generalization ability as well as the feasibility of binaural Transformers in sound localization. Furthermore, the analysis of the attention maps is provided to give additional insights on the interpretation of the localization process in a natural reverberant environment.

GCN-FFNN: A Two-Stream Deep Model for Learning Solution to Partial Differential Equations

This paper introduces a novel two-stream deep model based on graph convolutional network (GCN) architecture and feed-forward neural networks (FFNN) for learning the solution of nonlinear partial differential equations (PDEs). The model aims at incorporating both graph and grid input representations using two streams corresponding to GCN and FFNN models, respectively. Each stream layer receives and processes its own input representation. As opposed to FFNN which receives a grid-like structure, the GCN stream layer operates on graph input data where the neighborhood information is incorporated through the adjacency matrix of the graph. In this way, the proposed GCN-FFNN model learns from two types of input representations, i.e. grid and graph data, obtained via the discretization of the PDE domain. The GCN-FFNN model is trained in two phases. In the first phase, the model parameters of each stream are trained separately. Both streams employ the same error function to adjust their parameters by enforcing the models to satisfy the given PDE as well as its initial and boundary conditions on grid or graph collocation (training) data. In the second phase, the learned parameters of two-stream layers are frozen and their learned representation solutions are fed to fully connected layers whose parameters are learned using the previously used error function. The learned GCN-FFNN model is tested on test data located both inside and outside the PDE domain. The obtained numerical results demonstrate the applicability and efficiency of the proposed GCN-FFNN model over individual GCN and FFNN models on 1D-Burgers, 1D-Schr\"odinger, 2D-Burgers and 2D-Schr\"odinger equations.

AA-TransUNet: Attention Augmented TransUNet For Nowcasting Tasks

Data driven modeling based approaches have recently gained a lot of attention in many challenging meteorological applications including weather element forecasting. This paper introduces a novel data-driven predictive model based on TransUNet for precipitation nowcasting task. The TransUNet model which combines the Transformer and U-Net models has been previously successfully applied in medical segmentation tasks. Here, TransUNet is used as a core model and is further equipped with Convolutional Block Attention Modules (CBAM) and Depthwise-separable Convolution (DSC). The proposed Attention Augmented TransUNet (AA-TransUNet) model is evaluated on two distinct datasets: the Dutch precipitation map dataset and the French cloud cover dataset. The obtained results show that the proposed model outperforms other examined models on both tested datasets. Furthermore, the uncertainty analysis of the proposed AA-TransUNet is provided to give additional insights on its predictions.

Multistream Graph Attention Networks for Wind Speed Forecasting

Reliable and accurate wind speed prediction has significant impact in many industrial sectors such as economic, business and management among others. This paper presents a new model for wind speed prediction based on Graph Attention Networks (GAT). In particular, the proposed model extends GAT architecture by equipping it with a learnable adjacency matrix as well as incorporating a new attention mechanism with the aim of obtaining attention scores per weather variable. The output of the GAT based model is combined with the LSTM layer in order to exploit both the spatial and temporal characteristics of the multivariate multidimensional historical weather data. Real weather data collected from several cities in Denmark and Netherlands are used to conduct the experiments and evaluate the performance of the proposed model. We show that in comparison to previous architectures used for wind speed prediction, the proposed model is able to better learn the complex input-output relationships of the weather data. Furthermore, thanks to the learned attention weights, the model provides an additional insights on the most important weather variables and cities for the studied prediction task.

TENT: Tensorized Encoder Transformer for Temperature Forecasting

Reliable weather forecasting is of great importance in science, business and society. The best performing data-driven models for weather prediction tasks rely on recurrent or convolutional neural networks, where some of which incorporate attention mechanisms. In this work, we introduce a new model based on the Transformer architecture for weather forecasting. The proposed Tensorial Encoder Transformer (TENT) model is equipped with tensorial attention and thus it exploits the spatiotemporal structure of weather data by processing it in multidimensional tensorial format. We show that compared to the encoder part of the original transformer and 3D convolutional neural networks, the proposed TENT model can better model the underlying complex pattern of weather data for the studied temperature prediction task. Experiments on two real-life weather datasets are performed. The datasets consist of historical measurements from USA, Canada and European cities. The first dataset contains hourly measurements of weather attributes for 30 cities in USA and Canada from October 2012 to November 2017. The second dataset contains daily measurements of weather attributes of 18 cities across Europe from May 2005 to April 2020. We use attention scores calculated from our attention mechanism to shed light on the decision-making process of our model and have insight knowledge on the most important cities for the task.