Surface defect detection plays an increasingly important role in manufacturing industry to guarantee the product quality. Many deep learning methods have been widely used in surface defect detection tasks, and have been proven to perform well in defects classification and location. However, deep learning-based detection methods often require plenty of data for training, which fail to apply to the real industrial scenarios since the distribution of defect categories is often imbalanced. In other words, common defect classes have many samples but rare defect classes have extremely few samples, and it is difficult for these methods to well detect rare defect classes. To solve the imbalanced distribution problem, in this paper we propose TL-SDD: a novel Transfer Learning-based method for Surface Defect Detection. First, we adopt a two-phase training scheme to transfer the knowledge from common defect classes to rare defect classes. Second, we propose a novel Metric-based Surface Defect Detection (M-SDD) model. We design three modules for this model: (1) feature extraction module: containing feature fusion which combines high-level semantic information with low-level structural information. (2) feature reweighting module: transforming examples to a reweighting vector that indicates the importance of features. (3) distance metric module: learning a metric space in which defects are classified by computing distances to representations of each category. Finally, we validate the performance of our proposed method on a real dataset including surface defects of aluminum profiles. Compared to the baseline methods, the performance of our proposed method has improved by up to 11.98% for rare defect classes.

This paper aims to explain adversarial attacks in terms of how adversarial perturbations contribute to the attacking task. We estimate attributions of different image regions to the decrease of the attacking cost based on the Shapley value. We define and quantify interactions among adversarial perturbation pixels, and decompose the entire perturbation map into relatively independent perturbation components. The decomposition of the perturbation map shows that adversarially-trained DNNs have more perturbation components in the foreground than normally-trained DNNs. Moreover, compared to the normally-trained DNN, the adversarially-trained DNN have more components which mainly decrease the score of the true category. Above analyses provide new insights into the understanding of adversarial attacks.

Automated segmentation in medical image analysis is a challenging task that requires a large amount of manually labeled data. However, most existing learning-based approaches usually suffer from limited manually annotated medical data, which poses a major practical problem for accurate and robust medical image segmentation. In addition, most existing semi-supervised approaches are usually not robust compared with the supervised counterparts, and also lack explicit modeling of geometric structure and semantic information, both of which limit the segmentation accuracy. In this work, we present SimCVD, a simple contrastive distillation framework that significantly advances state-of-the-art voxel-wise representation learning. We first describe an unsupervised training strategy, which takes two views of an input volume and predicts their signed distance maps of object boundaries in a contrastive objective, with only two independent dropout as mask. This simple approach works surprisingly well, performing on the same level as previous fully supervised methods with much less labeled data. We hypothesize that dropout can be viewed as a minimal form of data augmentation and makes the network robust to representation collapse. Then, we propose to perform structural distillation by distilling pair-wise similarities. We evaluate SimCVD on two popular datasets: the Left Atrial Segmentation Challenge (LA) and the NIH pancreas CT dataset. The results on the LA dataset demonstrate that, in two types of labeled ratios (i.e., 20% and 10%), SimCVD achieves an average Dice score of 90.85% and 89.03% respectively, a 0.91% and 2.22% improvement compared to previous best results. Our method can be trained in an end-to-end fashion, showing the promise of utilizing SimCVD as a general framework for downstream tasks, such as medical image synthesis and registration.

Automatic speech emotion recognition (SER) is a challenging task that plays a crucial role in natural human-computer interaction. One of the main challenges in SER is data scarcity, i.e., insufficient amounts of carefully labeled data to build and fully explore complex deep learning models for emotion classification. This paper aims to address this challenge using a transfer learning strategy combined with spectrogram augmentation. Specifically, we propose a transfer learning approach that leverages a pre-trained residual network (ResNet) model including a statistics pooling layer from speaker recognition trained using large amounts of speaker-labeled data. The statistics pooling layer enables the model to efficiently process variable-length input, thereby eliminating the need for sequence truncation which is commonly used in SER systems. In addition, we adopt a spectrogram augmentation technique to generate additional training data samples by applying random time-frequency masks to log-mel spectrograms to mitigate overfitting and improve the generalization of emotion recognition models. We evaluate the effectiveness of our proposed approach on the interactive emotional dyadic motion capture (IEMOCAP) dataset. Experimental results indicate that the transfer learning and spectrogram augmentation approaches improve the SER performance, and when combined achieve state-of-the-art results.

Understanding complex social interactions among agents is a key challenge for trajectory prediction. Most existing methods consider the interactions between pairwise traffic agents or in a local area, while the nature of interactions is unlimited, involving an uncertain number of agents and non-local areas simultaneously. Besides, they treat heterogeneous traffic agents the same, namely those among agents of different categories, while neglecting people's diverse reaction patterns toward traffic agents in ifferent categories. To address these problems, we propose a simple yet effective Unlimited Neighborhood Interaction Network (UNIN), which predicts trajectories of heterogeneous agents in multiple categories. Specifically, the proposed unlimited neighborhood interaction module generates the fused-features of all agents involved in an interaction simultaneously, which is adaptive to any number of agents and any range of interaction area. Meanwhile, a hierarchical graph attention module is proposed to obtain category-to-category interaction and agent-to-agent interaction. Finally, parameters of a Gaussian Mixture Model are estimated for generating the future trajectories. Extensive experimental results on benchmark datasets demonstrate a significant performance improvement of our method over the state-of-the-art methods.

Washington [US], August 14 (ANI): Researchers have developed a deep learning model that is capable of classifying a brain tumour as one of six common types, using a single 3D MRI scan, during a new study. The study by researchers from the Washington University School of Medicine has been published in Radiology: Artificial Intelligence. "This is the first study to address the most common intracranial tumours and to directly determine the tumour class or the absence of tumour from a 3D MRI volume," said Satrajit Chakrabarty, M.S., a doctoral student under the direction of Aristeidis Sotiras, PhD, and Daniel Marcus, PhD, in Mallinckrodt Institute of Radiology's Computational Imaging Lab at Washington University School of Medicine in St. Louis, Missouri. The six most common intracranial tumour types are high-grade glioma, low-grade glioma, brain metastases, meningioma, pituitary adenoma and acoustic neuroma. Each was documented through histopathology, which requires surgically removing tissue from the site of suspected cancer and examining it under a microscope.

After the finish of the course of neural network, I just started another course which is "Improving deep neural networks: Hyperparameter tuning, Regularization and Optimization". In this course, it will focus on how the neural networks work better. Usually, hyperparameter tuning is quite useful but you cannot get the optimal setting at the first time. You have to iterate the whole idea by changing the code for experiment to obtain the best performance. Once you got a optimized hyperparameter but it cannot be transferred to other aspect and you have to work it out again.

We all now what is sequence i.e sequence is flow of alphabet's either individually or in string format in particular constant direction, Now it may be alphabet's or string it depends on what type Data you have. Sequence to Sequence is the same concept introduced in Deep Learning to work on long text/digit Data. In older days, when Deep Learning was not to popular there were problem . That Deep Learning model's were not able to store long context of Data. So, many of Reasearche's perform to solve this problem.

This Hedge Fund Stocks Package is designed for investors and analysts who need predictions of the best-performing stocks according to I Know First's stock market algorithm. Package Name: Hedge Fund Stocks Recommended Positions: Long Forecast Length: 1 Year (8/13/20 – 8/13/21) I Know First Average: 402.12% I Know First's State of the Art Algorithm accurately forecasted 10 out of 10 trades for the 1 Year time period. GME was our the best stock pick with a return of 3495.58%. The suggested trades for PDCE and URI also had notable 1 Year yields of 144.59% and 97.71%, respectively.

DALLAS – August 3, 2021 – Using artificial intelligence (AI), researchers from UT Southwestern have developed a way to accurately predict which skin cancers are highly metastatic. The findings, published as the July cover article of Cell Systems, show the potential for AI-based tools to revolutionize pathology for cancer and a variety of other diseases. "We now have a general framework that allows us to take tissue samples and predict mechanisms inside cells that drive disease, mechanisms that are currently inaccessible in any other way," said study leader Gaudenz Danuser, Ph.D., Professor and Chair of the Lyda Hill Department of Bioinformatics at UTSW. AI technology has significantly advanced over the past several years, Dr. Danuser explained, with deep learning-based methods able to distinguish minute differences in images that are essentially invisible to the human eye. Researchers have proposed using this latent information to look for differences in disease characteristics that could offer insight on prognoses or guide treatments.