Cost-Aware TVM (CAT) Tensorization for Modern Deep Learning Accelerators

This paper addresses the problem of rapid, precise, and efficient identification and diagnosis of grape leaf diseases by proposing the Deep-Broad Learning Network Model (ABLSS), which combines a Broad Learning network model with deep learning techniques. The model is optimized using the Adam algorithm based on BLS, and incorporates the LTM mechanism, which significantly enhances learning efficiency, stability, and recognition accuracy. Additionally, by integrating deep learning network optimization techniques, a SENet attention mechanism is added between the mapping and enhancement layers of BLS. Furthermore, based on the U-Net segmentation model, the method integrates dilated spatial pyramid pooling and feature pyramid networks. Dilated convolutions with varying dilation rates are used to capture multi-scale contextual information, which providing rich semantic information and high-resolution details during the decoding process. This improves the ABLSS model’s ability to identify small disease spots. Experimental results show that the ABLSS model achieves the highest recognition accuracy for three types of diseases with similar features on grape leaves, with an average accuracy improvement of 7.69% over BLS and 4.48% over deep learning networks. The MIOU of the segmentation model reaches 86.61%, which is a 6.48% improvement over the original U-Net model, and the MPA is 90.23%, a 8.09% improvement over the original U-Net. These results demonstrate that the proposed method significantly improves the algorithm’s recognition accuracy for small and irregular complex images. The ABLSS model recognizes images 0.375 seconds faster than the deep learning network, achieving a 72.12% speed improvement, thereby significantly enhancing the recognition efficiency of fine features. The ABLSS model combines the high recognition accuracy of deep learning with the fast processing speed of Broad Learning, while overcoming the limitations of BLS in recognizing complex images. This study provides valuable support for the development of smart orchard technologies and the optimization of learning network models.

Checking the kinship of facial images is a difficult research topic in computer vision that has attracted attention in recent years. The methods suggested so far are not strong enough to predict kinship relationships only by facial appearance. To mitigate this problem, we propose a new approach called Deep-Tensor+ELM to kinship verification based on deep (VGG-Face descriptor) and tensor (BSIF-Tensor & LPQ-Tensor using MSIDA method) features through Extreme Learning Machine (ELM). While ELM aims to deal with small size training features dimension, deep and tensor features are proven to provide significant enhancement over shallow features or vector-based counterparts. We evaluate our proposed method on the largest kinship benchmark namely FIW database using four Grandparent-Grandchild relations (GF-GD, GF-GS, GM-GD and GM-GS). The results obtained are positively compared with some modern methods, including those that rely on deep learning.

Introduction. The significance of machine learning under the conditions of digital transformation of industry, and methods of implementing deep learning to provide the performance of trust management systems are considered. The necessity of using convolutional artificial neural networks for deep machine learning is determined. Various technologies and architectures for the implementation of artificial neural networks are briefly considered; a comparative analysis of their performance is carried out. The work objective is to study the need to develop new approaches to the architecture of computing machines for solving problems of deep machine learning in the trust management system implementation.Materials and Methods. In the context of digital transformation, the use of artificial intelligence reaches a new level. The technical implementation of artificial neural systems with deep machine learning is based on the use of one of three basic technologies: high performance computing (HPC) with parallel data processing, neuromorphic computing (NC), and quantum computing (QC).Results. Implementation models for deep machine learning, basic technologies and architecture of computing machines, as well as requirements for trust assurance in control systems using deep machine learning are analyzed. The problem of shortage of computation power for solving such problems is identified. None of the currently existing technologies can solve the full range of learning and impedance problems. The current level of technology does not provide information security and reliability of neural networks. The practical implementation of trust management systems with deep machine learning based on existing technologies for a significant part of the tasks does not provide a sufficient level of performance.Discussion and Conclusions. The study made it possible to identify the challenge of the computation power shortage for solving problems of deep machine learning. Through the analysis of the requirements for trust management systems, the external challenges of their implementation on the basis of existing technologies, and the need to develop new approaches to the computer architecture are determined.

Detecting subtle signals from small earthquakes triggered by transient stresses from the surface waves of large magnitude earthquakes can contribute to a more general understanding of how earthquakes nucleate and interact with each other. However, searching for signals from such small earthquakes in thousands of seismograms is overwhelming, and discriminating them from a miscellany of noise is challenging. Here, we explore how we can automate the detection of such dynamically triggered earthquakes using a simple, diagnostic signal-to-noise ratio (SNR) threshold as well as a convolutional deep metric learning network. Our analysis shows that the deep learning network was more reliable at detecting small earthquakes than the SNR method.

Image segmentation is the process of delineating the regions occupied by objects of interest in a given image. This operation is a fundamentally required first step in numerous applications of medical imagery. In the medical imaging field, this activity has a rich literature that spans over 45 years. In spite of numerous advances, including deep learning (DL) networks (DLNs) in recent years, the problem has defied a robust, fail-safe, and satisfactory solution, especially for objects that are manifest with low contrast, are spatially sparse, have variable shape among individuals, or are affected by imaging artifacts, pathology, or post-treatment change in the body. Although image processing techniques, notably DLNs, are uncanny in their ability to harness low-level intensity pattern information on objects, they fall short in the high-level task of identifying and localizing an entire object as a gestalt. This dilemma has been a fundamental unmet challenge in medical image segmentation. In this paper, we demonstrate that by synergistically marrying the unmatched strengths of high-level human knowledge (i.e., natural intelligence (NI)) with the capabilities of DL networks (i.e., artificial intelligence (AI)) in garnering intricate details, these challenges can be significantly overcome. Focusing on the object recognition task, we formulate an anatomy-guided DL object recognition approach named Automatic Anatomy Recognition-Deep Learning (AAR-DL) which combines an advanced anatomy-modeling strategy, model-based non-DL object recognition, and DL object detection networks to achieve expert human-like performance.

Deep learning (also known as deep structured learning, hierarchical learning, or deep machine learning) is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and nonlinear transformations. Deep learning has been characterized as a class of machine learning algorithms with the following characteristics [257]: • They use a cascade of many layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The algorithms may be supervised or unsupervised and applications include pattern analysis (unsupervised) and classification (supervised). • They are based on the (unsupervised) learning of multiple levels of features or representations of the data. Higher-level features are derived from lower-level features to form a hierarchical representation. • They are part of the broader machine learning field of learning representations of data. • They learn multiple levels of representations that correspond to different levels of abstraction; the levels form a hierarchy of concepts. These definitions have in common: multiple layers of nonlinear processing units and the supervised or unsupervised learning of feature representations in each layer, with the layers forming a hierarchy from low-level to high-level features. Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks (DBN), and recurrent neural networks have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition, and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. In this chapter we start in Section 7.1 with an introduction, giving a brief history of this field, the relevant literature, and its applications. Then we study some basic concepts of deep learning such as convolutional neural networks, recurrent neural networks, backpropagation algorithm, restricted Boltzmann machines, and deep learning networks in Section 7.2. Then we illustrate three examples for Apache Spark implementation for mobile big data (MBD), user moving pattern extraction, and combination with nonparametric Bayesian learning, respectively, in Sections 7.3 through 7.5. Finally, we have summary in Section 7.6.

Knowledge-driven building extraction method exhibits a restricted adaptability scope and is vulnerable to external factors that affect its extraction accuracy. On the other hand, data-driven building extraction method lacks interpretability, heavily relies on extensive training data, and may result in extraction outcomes with building boundary blur issues. The integration of pre-existing knowledge with data-driven learning is essential for the intelligent identification and extraction of buildings from high-resolution aerial images. To overcome the limitations of current deep learning building extraction networks in effectively leveraging prior knowledge of aerial images, a geometric significance-aware deep mutual learning network (GSDMLNet) is proposed. Firstly, the GeoSay algorithm is utilized to derive building geometric significance feature maps as prior knowledge and integrate them into the deep learning network to enhance the targeted extraction of building features. Secondly, a bi-directional guidance attention module (BGAM) is developed to facilitate deep mutual learning between the building feature map and the building geometric significance feature map within the dual-branch network. Furthermore, the deployment of an enhanced flow alignment module (FAM++) is utilized to produce high-resolution, robust semantic feature maps with strong interpretability. Ultimately, a multi-objective loss function is crafted to refine the network’s performance. Experimental results demonstrate that the GSDMLNet excels in building extraction tasks within densely populated and diverse urban areas, reducing misidentification of shadow-obscured regions and color-similar terrains lacking building structural features. This approach effectively ensures the precise acquisition of urban building information in aerial images.

To construct the risk prediction model of acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and verify its effectiveness based on deep learning and back propagation algorithm neural network (BP neural network). Based on the relevant data of 1 326 patients with chronic obstructive pulmonary disease (COPD) in the team's previous clinical study, the acute exacerbation, and its risk factors during the stable period and 6 months of follow-up were recorded and analyzed. Combined with previous clinical research data and expert questionnaire results, the independent risk factors of AECOPD after screening and optimization by multivariate Logistic regression including gender, body mass index (BMI) classification, number of acute exacerbation, duration of acute exacerbation and forced expiratory volume in one second (FEV1) were used to build the BP neural network by Python 3.6 programming language and Tensorflow 1.12 deep learning framework. The patients were randomly selected according to the ratio of 4:1 to generate the training group and the test group, of which, the training group had 1 061 sample data while the test group had 265 pieces of sample data. The training group was used to establish the prediction model of neural network, and the test group was used for back-substitution test. When using the training group data to construct the neural network model, the training group was randomly divided into training set and verification set according to the ratio of 4:1. There were 849 training samples in the training set and 212 verification samples in the verification set. The optimal model was screened by adjusting the parameters of the neural network and combining the area under the receiver operator characteristic curve (AUC), and the sample data of the test group was substituted into the model for verification. The independent risk factors including gender, BMI classification, number of acute exacerbation, duration of acute exacerbation and FEV1 were collected from the team's previous clinical research, and the AECOPD risk prediction model was constructed based on deep learning and BP neural network. After 10 000 training sessions, the accuracy of the AECOPD risk prediction model in the validation set of the training group was 83.09%. When the number of training times reached 8 000, the accuracy basically tended to be stable and the prediction ability reached the upper limit. The AECOPD risk prediction model trained for 10 000 times was used to predict the risk of the validation set data, and the receiver operator characteristic curve (ROC curve) analysis showed that the AUC was 0.803. When using this model to predict the risk of the data of the test group, the accuracy rate was 81.69%. The risk prediction model based on deep learning and BP neural network has a medium level of prediction efficiency for acute exacerbation within 6 months in COPD patients, which can evaluate the risk of AECOPD and assist the clinic in making accurate treatment decisions.

A hardware-independent time estimation method for inference process of convolutional layers on GPU

Technical note: Impact of tissue section thickness on accuracy of cell classification with a deep learning network.

Radar data mining is the key module for signal analysis, where patterns hidden inside of signals are gradually available in the learning process and its superiority is significant for enhancing the security of the radar emitter classification (REC) system. Owing to the disadvantage that radio frequency fingerprinting (RFF) caused by the imperfection of emitter’s hardware is difficult to forge, current deep-learning REC methods based on deep-learning techniques, e.g., convolutional neural network (CNN) and long short term memory (LSTM) are difficult to capture the stable RFF features. In this paper, an online and non-cooperative multi-modal generic representation auxiliary learning REC model, namely muti-modal generic representation auxiliary learning networks (MGRALN), is put forward. Multi-modal means that multi-domain transformations are unified to a generic representation. After this, the representation is employed to facilitate mining the implicit information inside of the signals and to perform the better model robustness, which is achieved by using the available generic genenation to guide the network training and learning. Online means the learning process of REC is only once and the REC is end-to-end. Non-cooperative denotes no demodulation techniques are used before the REC task. Experimental results on the measured civil aviation radar data demonstrate that the proposed method enables one to achieve superior performance.

BackgroundTo determine if mammographic features from deep learning networks can be applied in breast cancer to identify groups at interval invasive cancer risk due to masking beyond using traditional breast density measures.MethodsFull-field digital screening mammograms acquired in our clinics between 2006 and 2015 were reviewed. Transfer learning of a deep learning network with weights initialized from ImageNet was performed to classify mammograms that were followed by an invasive interval or screen-detected cancer within 12 months of the mammogram. Hyperparameter optimization was performed and the network was visualized through saliency maps. Prediction loss and accuracy were calculated using this deep learning network. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values were generated with the outcome of interval cancer using the deep learning network and compared to predictions from conditional logistic regression with errors quantified through contingency tables.ResultsPre-cancer mammograms of 182 interval and 173 screen-detected cancers were split into training/test cases at an 80/20 ratio. Using Breast Imaging-Reporting and Data System (BI-RADS) density alone, the ability to correctly classify interval cancers was moderate (AUC = 0.65). The optimized deep learning model achieved an AUC of 0.82. Contingency table analysis showed the network was correctly classifying 75.2% of the mammograms and that incorrect classifications were slightly more common for the interval cancer mammograms. Saliency maps of each cancer case found that local information could highly drive classification of cases more than global image information.ConclusionsPre-cancerous mammograms contain imaging information beyond breast density that can be identified with deep learning networks to predict the probability of breast cancer detection.

Summary Training neural networks to detect features on seismic requires care from the interpreter to label all the true positives present on a training image. But often, the interpreter is solely interested in detecting features in a single interval. This narrowed focus arises from either a preference for zones associated with commercial hydrocarbon systems, or because the features in the zone of interest are characteristically different than similar features in overlying or underlying zones. In either case, it would be desirable to create a volumetrically consistent subdivision of stratal layers before feature labeling work begins in earnest. To achieve these ends, the authors introduce a novel way to perform stratal zonation using a Hierarchical Deep Learning (HDL) network. This approach simultaneously segments the entire seismic image into an arbitrary number of stratal zones using a multi-class network. Each stratal zone can then be further subdivided into a hierarchical arrangement of layers. Once trained, the HDL network’s inference can then be iteratively refined using an increasingly rich set of control points. On each successive iteration, the HDL network returns an inference that more closely approaches the geoscientist’s expert opinion.

A local feature tensor similarity based deep learning approach is proposed in this paper for 3D face recognition. Once a set of salient points on the 3D mesh are detected, three scale and rotation invariant features are extracted to represent local surface around each salient point. The local features of all the salient points are concatenated to produce a 3rd order feature tensor to represent a 3D face. Similarity of two 3D faces can thus be measured by a similarity tensor calculated using the two feature tensors. To address the unavailability of large 3D face samples, a feature tensor based data augmentation approach is proposed to augment the number of feature tensors. Experimental results show that the ResNet model trained using the augmented feature tensors achieves the best performance among state of the art competitors, i.e. 99.71% and 96.2% accuracy are achieved for Bosphorus and BU3DFE database, respectively.

Millimeter wave (mmWave) provides extremely high throughput owing to their high bandwidth utilization over higher frequencies. To compensate for the severe loss and attenuation, beamforming training is used to determine the optimal beam direction and thereby improve directional transmission power. However, the training overhead will be significantly increased with narrower beams, particularly under conventional exhaustive schemes. Therefore, we propose a learning-based adjustable beam number training (LABNT) scheme to intelligently and flexibly select beam training candidates considering different user mobility and hybrid line-of-sight (LOS) and non-line-of-sight (NLOS) scenarios. In LABNT, several deep learning networks are parallelly constructed and connected to a reinforcement learning network to determine training candidates dynamically based on performance rewards. A novel enhanced feature selection method, that is, uniformly distributed mutual information, is developed based on the correlation of historical results to select the beam inputs for each deep learning network. In simulations, the proposed LABNT outperforms other existing schemes in terms of beam alignment accuracy, training latency, and system throughput. Moreover, following the IEEE 802.11ad/ay protocols, we implement realistic mmWave beamforming training on a programmable hardware platform with integrated 60 GHz wireless-gigabit (WiGig) transceiver devices. The experimental results on the WiGig platform demonstrate that the proposed LABNT scheme can achieve real-time performance with approximately Gbps transmission throughput and millisecond-level training overhead.