Spatial and Feature-selective Attention Have Distinct, Interacting Effects on Population-level Tuning

Attention can be deployed in different ways: When searching for a taxi in New York City, we can decide where to attend (e.g., to the street) and what to attend to (e.g., yellow cars). Although we use the same word to describe both processes, nonhuman primate data suggest that these produce distinct effects on neural tuning. This has been challenging to assess in humans, but here we used an opportunity afforded by multivariate decoding of MEG data. We found that attending to an object at a particular location and attending to a particular object feature produced effects that interacted multiplicatively. The two types of attention induced distinct patterns of enhancement in occipital cortex, with feature-selective attention producing relatively more enhancement of small feature differences and spatial attention producing relatively larger effects for larger feature differences. An information flow analysis further showed that stimulus representations in occipital cortex were Granger-caused by coding in frontal cortices earlier in time and that the timing of this feedback matched the onset of attention effects. The data suggest that spatial and feature-selective attention rely on distinct neural mechanisms that arise from frontal-occipital information exchange, interacting multiplicatively to selectively enhance task-relevant information.

Improved Mask R-CNN for Aircraft Detection in Remote Sensing Images

In recent years, remote sensing images has become one of the most popular directions in image processing. A small feature gap exists between satellite and natural images. Therefore, deep learning algorithms could be applied to recognize remote sensing images. We propose an improved Mask R-CNN model, called SCMask R-CNN, to enhance the detection effect in the high-resolution remote sensing images which contain the dense targets and complex background. Our model can perform object recognition and segmentation in parallel. This model uses a modified SC-conv based on the ResNet101 backbone network to obtain more discriminative feature information and adds a set of dilated convolutions with a specific size to improve the instance segmentation effect. We construct WFA-1400 based on the DOTA dataset because of the shortage of remote sensing mask datasets. We compare the improved algorithm with other state-of-the-art algorithms. The object detection AP50 and AP increased by 1–2% and 1%, respectively, objectively proving the effectiveness and the feasibility of the improved model.

Improved SSD-assisted algorithm for surface defect detection of electromagnetic luminescence

Defect detection of electromagnetic luminescence (EL) cells is the core step in the production and preparation of solar cell modules to ensure conversion efficiency and long service life of batteries. However, due to the lack of feature extraction capability for small feature defects, the traditional single shot multibox detector (SSD) algorithm performs not well in EL defect detection with high accuracy. Consequently, an improved SSD algorithm with modification in feature fusion in the framework of deep learning is proposed to improve the recognition rate of EL multi-class defects. A dataset containing images with four different types of defects through rotation, denoising, and binarization is established for the EL. The proposed algorithm can greatly improve the detection accuracy of the small-scale defect with the idea of feature pyramid networks. An experimental study on the detection of the EL defects shows the effectiveness of the proposed algorithm. Moreover, a comparison study shows the proposed method outperforms other traditional detection methods, such as the SIFT, Faster R-CNN, and YOLOv3, in detecting the EL defect.

Novel Memory Structures in QCA Nano Technology

Quantum-dot Cellular Automata (QCA) is a new emerging technology for designing electronic circuits in nanoscale. QCA technology comes to overcome the CMOS limitation and to be a good alternative as it can work in ultra-high-speed. QCA brought researchers attention due to many features such as low power consumption, small feature size in addition to high frequency. Designing circuits in QCA technology with minimum costs such as cells count and the area is very important. This paper presents novel structures of D-latch and D-Flip Flop with the lower area and cell count. The proposed Flip-Flop has SET and RESET ability. The proposed latch and Flip-Flop have lower complexity compared with counterparts in terms of cell counts by 32% and 26% respectively. The proposed circuits are designed and simulated in QCADesigner software.

On the Interpretability of Machine Learning Models and Experimental Feature Selection in Case of Multicollinear Data

In the field of machine learning, a considerable amount of research is involved in the interpretability of models and their decisions. The interpretability contradicts the model quality. Random Forests are among the best quality technologies of machine learning, but their operation is of “black box” character. Among the quantifiable approaches to the model interpretation, there are measures of association of predictors and response. In case of the Random Forests, this approach usually consists of calculating the model’s feature importances. Known methods, including the built-in one, are less suitable in settings with strong multicollinearity of features. Therefore, we propose an experimental approach to the feature selection task, a greedy forward feature selection method with least-trees-used criterion. It yields a set of most informative features that can be used in a machine learning (ML) training process with similar prediction quality as the original feature set. We verify the results of the proposed method on two known datasets, one with small feature multicollinearity and another with large feature multicollinearity. The proposed method also allows for a domain expert help with selecting among equally important features, which is known as the human-in-the-loop approach.

Additive Manufactured Polymer-Bonded Isotropic NdFeB Magnets by Stereolithography and Their Comparison to Fused Filament Fabricated and Selective Laser Sintered Magnets

Magnetic isotropic NdFeB powder with a spherical morphology is used to 3D print magnets by stereolithography (SLA). Complex magnets with small feature sizes in a superior surface quality can be printed with SLA. The magnetic properties of the 3D printed bonded magnets are investigated and compared with magnets manufactured by fused filament fabrication (FFF), and selective laser sintering (SLS). All methods use the same hard magnetic isotropic NdFeB powder material. FFF and SLA use a polymer matrix material as binder, SLS sinters the powder directly. SLA can print magnets with a remanence of 388 mT and a coercivity of 0.923 T. A complex magnetic design for speed wheel sensing applications is presented and printed with all methods.

Where Will You Park? Predicting Vehicle Locations for Vehicle-to-Grid

Vehicle-to-grid services draw power or curtail demand from electric vehicles when they are connected to a compatible charging station. In this paper, we investigated automated machine learning for predicting when vehicles are likely to make such a connection. Using historical data collected from a vehicle tracking service, we assessed the technique’s ability to learn and predict when a fleet of 48 vehicles was parked close to charging stations and compared this with two moving average techniques. We found the ability of all three approaches to predict when individual vehicles could potentially connect to charging stations to be comparable, resulting in the same set of 30 vehicles identified as good candidates to participate in a vehicle-to-grid service. We concluded that this was due to the relatively small feature set and that machine learning techniques were likely to outperform averaging techniques for more complex feature sets. We also explored the ability of the approaches to predict total vehicle availability and found that automated machine learning achieved the best performance with an accuracy of 91.4%. Such technology would be of value to vehicle-to-grid aggregation services.

Investigating Feature Selection and Random Forests for Inter-Patient Heartbeat Classification

Finding an optimal combination of features and classifier is still an open problem in the development of automatic heartbeat classification systems, especially when applications that involve resource-constrained devices are considered. In this paper, a novel study of the selection of informative features and the use of a random forest classifier while following the recommendations of the Association for the Advancement of Medical Instrumentation (AAMI) and an inter-patient division of datasets is presented. Features were selected using a filter method based on the mutual information ranking criterion on the training set. Results showed that normalized beat-to-beat (R–R) intervals and features relative to the width of the ventricular depolarization waves (QRS complex) are the most discriminative among those considered. The best results achieved on the MIT-BIH Arrhythmia Database were an overall accuracy of 96.14% and F1-scores of 97.97%, 73.06%, and 90.85% in the classification of normal beats, supraventricular ectopic beats, and ventricular ectopic beats, respectively. In comparison with other state-of-the-art approaches tested under similar constraints, this work represents one of the highest performances reported to date while relying on a very small feature vector.

Investigating Feature Selection and Random Forests for Inter-patient Heartbeat Classification

Finding an optimal combination of features and classifier is still an open problem in the development of automatic heartbeat classification systems, especially when applications that involve resource-constrained devices are considered. In this paper, a novel study of the selection of informative features and the use of a random forest classifier while following the recommendations of the Association for the Advancement of Medical Instrumentation (AAMI) and an inter-patient division of datasets is presented. Features were selected using a filter method based on the mutual information ranking criterion on the training set. Results showed that normalized R-R intervals and features relative to the width of the QRS complex are the most discriminative among those considered. The best results achieved on the MIT-BIH Arrhythmia Database were an overall accuracy of 96.14% and F1-scores of 97.97%, 73.06%, and 90.85% in the classification of normal beats, supraventricular ectopic beats, and ventricular ectopic beats respectively. In comparison with other state of the art approaches tested under similar constraints, this work represents one of the highest performances reported to date while relying on a very small feature vector.