Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userMulti-channel coronal hole detection with convolutional neural networks
74
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We develop a reliable, fully automatic method for the detection of coronal holes, that provides consistent full-disk segmentation maps over the full solar cycle and can perform in real-time. We use a convolutional neural network to identify the boundaries of coronal holes from the seven EUV channels of the Atmospheric Imaging Assembly (AIA) as well as from line-of-sight magnetograms from the Helioseismic and Magnetic Imager (HMI) onboard the Solar Dynamics Observatory (SDO). For our primary model (Coronal Hole RecOgnition Neural Network Over multi-Spectral-data; CHRONNOS) we use a progressively growing network approach that allows for efficient training, provides detailed segmentation maps and takes relations across the full solar-disk into account. We provide a thorough evaluation for performance, reliability and consistency by comparing the model results to an independent manually curated test set. Our model shows good agreement to the manual labels with an intersection-over-union (IoU) of 0.63. From the total of 261 coronal holes with an area $>1.5cdot10^{10}$ km$^2$ identified during the time range 11/2010 - 12/2016, 98.1% were correctly detected by our model. The evaluation over almost the full solar cycle no. 24 shows that our model provides reliable coronal hole detections, independent of the level of solar activity. From the direct comparison over short time scales of days to weeks, we find that our model exceeds human performance in terms of consistency and reliability. In addition, we train our model to identify coronal holes from each channel separately and show that the neural network provides the best performance with the combined channel information, but that coronal hole segmentation maps can be also obtained solely from line-of-sight magnetograms.
rate research
Read More
Graph neural networks (GNN) has been demonstrated to be effective in classifying graph structures. To further improve the graph representation learning ability, hierarchical GNN has been explored. It leverages the differentiable pooling to cluster nodes into fixed groups, and generates a coarse-grained structure accompanied with the shrinking of the original graph. However, such clustering would discard some graph information and achieve the suboptimal results. It is because the node inherently has different characteristics or roles, and two non-isomorphic graphs may have the same coarse-grained structure that cannot be distinguished after pooling. To compensate the loss caused by coarse-grained clustering and further advance GNN, we propose a multi-channel graph convolutional networks (MuchGCN). It is motivated by the convolutional neural networks, at which a series of channels are encoded to preserve the comprehensive characteristics of the input image. Thus, we define the specific graph convolutions to learn a series of graph channels at each layer, and pool graphs iteratively to encode the hierarchical structures. Experiments have been carefully carried out to demonstrate the superiority of MuchGCN over the state-of-the-art graph classification algorithms.
We devise an autoencoder based strategy to facilitate anomaly detection for boosted jets, employing Graph Neural Networks (GNNs) to do so. To overcome known limitations of GNN autoencoders, we design a symmetric decoder capable of simultaneously reconstructing edge features and node features. Focusing on latent space based discriminators, we find that such setups provide a promising avenue to isolate new physics and competing SM signatures from sensitivity-limiting QCD jet contributions. We demonstrate the flexibility and broad applicability of this approach using examples of $W$ bosons, top quarks, and exotic hadronically-decaying exotic scalar bosons.
Neural Architecture Search (NAS) has shifted network design from using human intuition to leveraging search algorithms guided by evaluation metrics. We study channel size optimization in convolutional neural networks (CNN) and identify the role it plays in model accuracy and complexity. Current channel size selection methods are generally limited by discrete sample spaces while suffering from manual iteration and simple heuristics. To solve this, we introduce an efficient dynamic scaling algorithm -- CONet -- that automatically optimizes channel sizes across network layers for a given CNN. Two metrics -- ``textit{Rank} and textit{Rank Average Slope} -- are introduced to identify the information accumulated in training. The algorithm dynamically scales channel sizes up or down over a fixed searching phase. We conduct experiments on CIFAR10/100 and ImageNet datasets and show that CONet can find efficient and accurate architectures searched in ResNet, DARTS, and DARTS+ spaces that outperform their baseline models.
We develop an algorithm which exceeds the performance of board certified cardiologists in detecting a wide range of heart arrhythmias from electrocardiograms recorded with a single-lead wearable monitor. We build a dataset with more than 500 times the number of unique patients than previously studied corpora. On this dataset, we train a 34-layer convolutional neural network which maps a sequence of ECG samples to a sequence of rhythm classes. Committees of board-certified cardiologists annotate a gold standard test set on which we compare the performance of our model to that of 6 other individual cardiologists. We exceed the average cardiologist performance in both recall (sensitivity) and precision (positive predictive value).
Spreadsheet table detection is the task of detecting all tables on a given sheet and locating their respective ranges. Automatic table detection is a key enabling technique and an initial step in spreadsheet data intelligence. However, the detection task is challenged by the diversity of table structures and table layouts on the spreadsheet. Considering the analogy between a cell matrix as spreadsheet and a pixel matrix as image, and encouraged by the successful application of Convolutional Neural Networks (CNN) in computer vision, we have developed TableSense, a novel end-to-end framework for spreadsheet table detection. First, we devise an effective cell featurization scheme to better leverage the rich information in each cell; second, we develop an enhanced convolutional neural network model for table detection to meet the domain-specific requirement on precise table boundary detection; third, we propose an effective uncertainty metric to guide an active learning based smart sampling algorithm, which enables the efficient build-up of a training dataset with 22,176 tables on 10,220 sheets with broad coverage of diverse table structures and layouts. Our evaluation shows that TableSense is highly effective with 91.3% recall and 86.5% precision in EoB-2 metric, a significant improvement over both the current detection algorithm that are used in commodity spreadsheet tools and state-of-the-art convolutional neural networks in computer vision.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
