PyTorchKR Papers

최신 논문

총 8,922편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2204.11786

Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications

Deep neural networks (DNNs) have achieved unprecedented success in the field of artificial intelligence (AI), including computer vision, natural language processing and speech recognition. However, their superior performance comes at the considerable cost of computational complexity, which greatly hinders their applications in many resource-constrained devices, such as mobile phones and Internet of Things (IoT) devices. Therefore, methods and techniques that are able to lift the efficiency bottleneck while preserving the high accuracy of DNNs are in great demand in order to enable numerous edge AI applications. This paper provides an overview of efficient deep learning methods, systems and applications. We start from introducing popular model compression methods, including pruning, factorization, quantization as well as compact model design. To reduce the large design cost of these manual solutions, we discuss the AutoML framework for each of them, such as neural architecture search (NAS) and automated pruning and quantization. We then cover efficient on-device training to enable user customization based on the local data on mobile devices. Apart from general acceleration techniques, we also showcase several task-specific accelerations for point cloud, video and natural language processing by exploiting their spatial sparsity and temporal/token redundancy. Finally, to support all these algorithmic advancements, we introduce the efficient deep learning system design from both software and hardware perspectives.

arXiv:2204.11751공식 코드 · Papers with Code

Adversarial Attention for Human Motion Synthesis

Analysing human motions is a core topic of interest for many disciplines, from Human-Computer Interaction, to entertainment, Virtual Reality and healthcare. Deep learning has achieved impressive results in capturing human pose in real-time. On the other hand, due to high inter-subject variability, human motion analysis models often suffer from not being able to generalise to data from unseen subjects due to very limited specialised datasets available in fields such as healthcare. However, acquiring human motion datasets is highly time-consuming, challenging, and expensive. Hence, human motion synthesis is a crucial research problem within deep learning and computer vision. We present a novel method for controllable human motion synthesis by applying attention-based probabilistic deep adversarial models with end-to-end training. We show that we can generate synthetic human motion over both short- and long-time horizons through the use of adversarial attention. Furthermore, we show that we can improve the classification performance of deep learning models in cases where there is inadequate real data, by supplementing existing datasets with synthetic motions.

arXiv:2204.11341

Deep Learning for Medical Image Registration: A Comprehensive Review

Image registration is a critical component in the applications of various medical image analyses. In recent years, there has been a tremendous surge in the development of deep learning (DL)-based medical image registration models. This paper provides a comprehensive review of medical image registration. Firstly, a discussion is provided for supervised registration categories, for example, fully supervised, dual supervised, and weakly supervised registration. Next, similarity-based as well as generative adversarial network (GAN)-based registration are presented as part of unsupervised registration. Deep iterative registration is then described with emphasis on deep similarity-based and reinforcement learning-based registration. Moreover, the application areas of medical image registration are reviewed. This review focuses on monomodal and multimodal registration and associated imaging, for instance, X-ray, CT scan, ultrasound, and MRI. The existing challenges are highlighted in this review, where it is shown that a major challenge is the absence of a training dataset with known transformations. Finally, a discussion is provided on the promising future research areas in the field of DL-based medical image registration.

arXiv:2204.11294

Colorectal cancer survival prediction using deep distribution based multiple-instance learning

Several deep learning algorithms have been developed to predict survival of cancer patients using whole slide images (WSIs).However, identification of image phenotypes within the WSIs that are relevant to patient survival and disease progression is difficult for both clinicians, and deep learning algorithms. Most deep learning based Multiple Instance Learning (MIL) algorithms for survival prediction use either top instances (e.g., maxpooling) or top/bottom instances (e.g., MesoNet) to identify image phenotypes. In this study, we hypothesize that wholistic information of the distribution of the patch scores within a WSI can predict the cancer survival better. We developed a distribution based multiple-instance survival learning algorithm (DeepDisMISL) to validate this hypothesis. We designed and executed experiments using two large international colorectal cancer WSIs datasets - MCO CRC and TCGA COAD-READ. Our results suggest that the more information about the distribution of the patch scores for a WSI, the better is the prediction performance. Including multiple neighborhood instances around each selected distribution location (e.g., percentiles) could further improve the prediction. DeepDisMISL demonstrated superior predictive ability compared to other recently published, state-of-the-art algorithms. Furthermore, our algorithm is interpretable and could assist in understanding the relationship between cancer morphological phenotypes and patients cancer survival risk.

arXiv:2204.10959공식 코드 · Papers with Code

A benchmark dataset for deep learning-based airplane detection: HRPlanes

Airplane detection from satellite imagery is a challenging task due to the complex backgrounds in the images and differences in data acquisition conditions caused by the sensor geometry and atmospheric effects. Deep learning methods provide reliable and accurate solutions for automatic detection of airplanes; however, huge amount of training data is required to obtain promising results. In this study, we create a novel airplane detection dataset called High Resolution Planes (HRPlanes) by using images from Google Earth (GE) and labeling the bounding box of each plane on the images. HRPlanes include GE images of several different airports across the world to represent a variety of landscape, seasonal and satellite geometry conditions obtained from different satellites. We evaluated our dataset with two widely used object detection methods namely YOLOv4 and Faster R-CNN. Our preliminary results show that the proposed dataset can be a valuable data source and benchmark data set for future applications. Moreover, proposed architectures and results of this study could be used for transfer learning of different datasets and models for airplane detection.

arXiv:2204.11669공식 코드 · Papers with Code

Deep-learning-enabled Brain Hemodynamic Mapping Using Resting-state fMRI

Cerebrovascular disease is a leading cause of death globally. Prevention and early intervention are known to be the most effective forms of its management. Non-invasive imaging methods hold great promises for early stratification, but at present lack the sensitivity for personalized prognosis. Resting-state functional magnetic resonance imaging (rs-fMRI), a powerful tool previously used for mapping neural activity, is available in most hospitals. Here we show that rs-fMRI can be used to map cerebral hemodynamic function and delineate impairment. By exploiting time variations in breathing pattern during rs-fMRI, deep learning enables reproducible mapping of cerebrovascular reactivity (CVR) and bolus arrive time (BAT) of the human brain using resting-state CO2 fluctuations as a natural 'contrast media'. The deep-learning network was trained with CVR and BAT maps obtained with a reference method of CO2-inhalation MRI, which included data from young and older healthy subjects and patients with Moyamoya disease and brain tumors. We demonstrate the performance of deep-learning cerebrovascular mapping in the detection of vascular abnormalities, evaluation of revascularization effects, and vascular alterations in normal aging. In addition, cerebrovascular maps obtained with the proposed method exhibited excellent reproducibility in both healthy volunteers and stroke patients. Deep-learning resting-state vascular imaging has the potential to become a useful tool in clinical cerebrovascular imaging.

arXiv:2204.10533

Fourier Imager Network (FIN): A deep neural network for hologram reconstruction with superior external generalization

Deep learning-based image reconstruction methods have achieved remarkable success in phase recovery and holographic imaging. However, the generalization of their image reconstruction performance to new types of samples never seen by the network remains a challenge. Here we introduce a deep learning framework, termed Fourier Imager Network (FIN), that can perform end-to-end phase recovery and image reconstruction from raw holograms of new types of samples, exhibiting unprecedented success in external generalization. FIN architecture is based on spatial Fourier transform modules that process the spatial frequencies of its inputs using learnable filters and a global receptive field. Compared with existing convolutional deep neural networks used for hologram reconstruction, FIN exhibits superior generalization to new types of samples, while also being much faster in its image inference speed, completing the hologram reconstruction task in ~0.04 s per 1 mm^2 of the sample area. We experimentally validated the performance of FIN by training it using human lung tissue samples and blindly testing it on human prostate, salivary gland tissue and Pap smear samples, proving its superior external generalization and image reconstruction speed. Beyond holographic microscopy and quantitative phase imaging, FIN and the underlying neural network architecture might open up various new opportunities to design broadly generalizable deep learning models in computational imaging and machine vision fields.

arXiv:2204.10713공식 코드 · Papers with Code

EmbedTrack -- Simultaneous Cell Segmentation and Tracking Through Learning Offsets and Clustering Bandwidths

A systematic analysis of the cell behavior requires automated approaches for cell segmentation and tracking. While deep learning has been successfully applied for the task of cell segmentation, there are few approaches for simultaneous cell segmentation and tracking using deep learning. Here, we present EmbedTrack, a single convolutional neural network for simultaneous cell segmentation and tracking which predicts easy to interpret embeddings. As embeddings, offsets of cell pixels to their cell center and bandwidths are learned. We benchmark our approach on nine 2D data sets from the Cell Tracking Challenge, where our approach performs on seven out of nine data sets within the top 3 contestants including three top 1 performances. The source code is publicly available at https://git.scc.kit.edu/kit-loe-ge/embedtrack.

arXiv:2204.09738

Res-CNN-BiLSTM Network for overcoming Mental Health Disturbances caused due to Cyberbullying through Social Media

Mental Health Disturbance has many reasons and cyberbullying is one of the major causes that does exploitation using social media as an instrument. The cyberbullying is done on the basis of Religion, Ethnicity, Age and Gender which is a sensitive psychological issue. This can be addressed using Natural Language Processing with Deep Learning, since social media is the medium and it generates massive form of data in textual form. Such data can be leveraged to find the semantics and derive what type of cyberbullying is done and who are the people involved for early measures. Since deriving semantics is essential we proposed a Hybrid Deep Learning Model named 1-Dimensional CNN-Bidirectional-LSTMs with Residuals shortly known as Res-CNN-BiLSTM. In this paper we have proposed the architecture and compared its performance with different approaches of Embedding Deep Learning Algorithms.

arXiv:2204.10277

Deep learning techniques for energy clustering in the CMS ECAL

The reconstruction of electrons and photons in CMS depends on topological clustering of the energy deposited by an incident particle in different crystals of the electromagnetic calorimeter (ECAL). These clusters are formed by aggregating neighbouring crystals according to the expected topology of an electromagnetic shower in the ECAL. The presence of upstream material (beampipe, tracker and support structures) causes electrons and photons to start showering before reaching the calorimeter. This effect, combined with the 3.8T CMS magnetic field, leads to energy being spread in several clusters around the primary one. It is essential to recover the energy contained in these satellite clusters in order to achieve the best possible energy resolution for physics analyses. Historically satellite clusters have been associated to the primary cluster using a purely topological algorithm which does not attempt to remove spurious energy deposits from additional pileup interactions (PU). The performance of this algorithm is expected to degrade during LHC Run 3 (2022+) because of the larger average PU levels and the increasing levels of noise due to the ageing of the ECAL detector. New methods are being investigated that exploit state-of-the-art deep learning architectures like Graph Neural Networks (GNN) and self-attention algorithms. These more sophisticated models improve the energy collection and are more resilient to PU and noise, helping to preserve the electron and photon energy resolution achieved during LHC Runs 1 and 2. This work will cover the challenges of training the models as well the opportunity that this new approach offers to unify the ECAL energy measurement with the particle identification steps used in the global CMS photon and electron reconstruction.

arXiv:2204.09676

Spatially-Preserving Flattening for Location-Aware Classification of Findings in Chest X-Rays

Chest X-rays have become the focus of vigorous deep learning research in recent years due to the availability of large labeled datasets. While classification of anomalous findings is now possible, ensuring that they are correctly localized still remains challenging, as this requires recognition of anomalies within anatomical regions. Existing deep learning networks for fine-grained anomaly classification learn location-specific findings using architectures where the location and spatial contiguity information is lost during the flattening step before classification. In this paper, we present a new spatially preserving deep learning network that preserves location and shape information through auto-encoding of feature maps during flattening. The feature maps, auto-encoder and classifier are then trained in an end-to-end fashion to enable location aware classification of findings in chest X-rays. Results are shown on a large multi-hospital chest X-ray dataset indicating a significant improvement in the quality of finding classification over state-of-the-art methods.

arXiv:2204.09664

Deep Learning meets Nonparametric Regression: Are Weight-Decayed DNNs Locally Adaptive?

We study the theory of neural network (NN) from the lens of classical nonparametric regression problems with a focus on NN's ability to adaptively estimate functions with heterogeneous smoothness -- a property of functions in Besov or Bounded Variation (BV) classes. Existing work on this problem requires tuning the NN architecture based on the function spaces and sample size. We consider a "Parallel NN" variant of deep ReLU networks and show that the standard $\ell_2$ regularization is equivalent to promoting the $\ell_p$-sparsity ($0<p<1$) in the coefficient vector of an end-to-end learned function bases, i.e., a dictionary. Using this equivalence, we further establish that by tuning only the regularization factor, such parallel NN achieves an estimation error arbitrarily close to the minimax rates for both the Besov and BV classes. Notably, it gets exponentially closer to minimax optimal as the NN gets deeper. Our research sheds new lights on why depth matters and how NNs are more powerful than kernel methods.

arXiv:2204.09578

Restructuring TCAD System: Teaching Traditional TCAD New Tricks

Traditional TCAD simulation has succeeded in predicting and optimizing the device performance; however, it still faces a massive challenge - a high computational cost. There have been many attempts to replace TCAD with deep learning, but it has not yet been completely replaced. This paper presents a novel algorithm restructuring the traditional TCAD system. The proposed algorithm predicts three-dimensional (3-D) TCAD simulation in real-time while capturing a variance, enables deep learning and TCAD to complement each other, and fully resolves convergence errors.

arXiv:2204.09050

House Price Prediction Based On Deep Learning

Since ancient times, what Chinese people have been pursuing is very simple, which is nothing more than "to live and work happily, to eat and dress comfortable". Today, more than 40 years after the reform and opening, people have basically solved the problem of food and clothing, and the urgent problem is housing. Nowadays, due to the storm of long-term rental apartment intermediary platforms such as eggshell, increasing the sense of insecurity of renters, as well as the urbanization in recent years and the scramble for people in major cities, this will make the future real estate market competition more intense. In order to better grasp the real estate price, let consumers buy a house reasonably, and provide a reference for the government to formulate policies, this paper summarizes the existing methods of house price prediction and proposes a house price prediction method based on mixed depth vision and text features.

arXiv:2204.09343

Utilizing unsupervised learning to improve sward content prediction and herbage mass estimation

Sward species composition estimation is a tedious one. Herbage must be collected in the field, manually separated into components, dried and weighed to estimate species composition. Deep learning approaches using neural networks have been used in previous work to propose faster and more cost efficient alternatives to this process by estimating the biomass information from a picture of an area of pasture alone. Deep learning approaches have, however, struggled to generalize to distant geographical locations and necessitated further data collection to retrain and perform optimally in different climates. In this work, we enhance the deep learning solution by reducing the need for ground-truthed (GT) images when training the neural network. We demonstrate how unsupervised contrastive learning can be used in the sward composition prediction problem and compare with the state-of-the-art on the publicly available GrassClover dataset collected in Denmark as well as a more recent dataset from Ireland where we tackle herbage mass and height estimation.

arXiv:2204.09051공식 코드 · Papers with Code

PR-DAD: Phase Retrieval Using Deep Auto-Decoders

Phase retrieval is a well known ill-posed inverse problem where one tries to recover images given only the magnitude values of their Fourier transform as input. In recent years, new algorithms based on deep learning have been proposed, providing breakthrough results that surpass the results of the classical methods. In this work we provide a novel deep learning architecture PR-DAD (Phase Retrieval Using Deep Auto- Decoders), whose components are carefully designed based on mathematical modeling of the phase retrieval problem. The architecture provides experimental results that surpass all current results.

arXiv:2204.08987

Deep learning based closed-loop optimization of geothermal reservoir production

To maximize the economic benefits of geothermal energy production, it is essential to optimize geothermal reservoir management strategies, in which geologic uncertainty should be considered. In this work, we propose a closed-loop optimization framework, based on deep learning surrogates, for the well control optimization of geothermal reservoirs. In this framework, we construct a hybrid convolution-recurrent neural network surrogate, which combines the convolution neural network (CNN) and long short-term memory (LSTM) recurrent network. The convolution structure can extract spatial information of geologic parameter fields and the recurrent structure can approximate sequence-to-sequence mapping. The trained model can predict time-varying production responses (rate, temperature, etc.) for cases with different permeability fields and well control sequences. In the closed-loop optimization framework, production optimization based on the differential evolution (DE) algorithm, and data assimilation based on the iterative ensemble smoother (IES), are performed alternately to achieve real-time well control optimization and geologic parameter estimation as the production proceeds. In addition, the averaged objective function over the ensemble of geologic parameter estimations is adopted to consider geologic uncertainty in the optimization process. Several geothermal reservoir development cases are designed to test the performance of the proposed production optimization framework. The results show that the proposed framework can achieve efficient and effective real-time optimization and data assimilation in the geothermal reservoir production process.

arXiv:2204.08499공식 코드 · Papers with Code

DeepCore: A Comprehensive Library for Coreset Selection in Deep Learning

Coreset selection, which aims to select a subset of the most informative training samples, is a long-standing learning problem that can benefit many downstream tasks such as data-efficient learning, continual learning, neural architecture search, active learning, etc. However, many existing coreset selection methods are not designed for deep learning, which may have high complexity and poor generalization performance. In addition, the recently proposed methods are evaluated on models, datasets, and settings of different complexities. To advance the research of coreset selection in deep learning, we contribute a comprehensive code library, namely DeepCore, and provide an empirical study on popular coreset selection methods on CIFAR10 and ImageNet datasets. Extensive experiments on CIFAR10 and ImageNet datasets verify that, although various methods have advantages in certain experiment settings, random selection is still a strong baseline.

arXiv:2204.08989공식 코드 · Papers with Code

Efficient Deep Learning-based Estimation of the Vital Signs on Smartphones

With the increasing use of smartphones in our daily lives, these devices have become capable of performing many complex tasks. Concerning the need for continuous monitoring of vital signs, especially for the elderly or those with certain types of diseases, the development of algorithms that can estimate vital signs using smartphones has attracted researchers worldwide. In particular, researchers have been exploring ways to estimate vital signs, such as heart rate, oxygen saturation levels, and respiratory rate, using algorithms that can be run on smartphones. However, many of these algorithms require multiple pre-processing steps that might introduce some implementation overheads or require the design of a couple of hand-crafted stages to obtain an optimal result. To address this issue, this research proposes a novel end-to-end solution to mobile-based vital sign estimation using deep learning that eliminates the need for pre-processing. By using a fully convolutional architecture, the proposed model has much fewer parameters and less computational complexity compared to the architectures that use fully-connected layers as the prediction heads. This also reduces the risk of overfitting. Additionally, a public dataset for vital sign estimation, which includes 62 videos collected from 35 men and 27 women, is provided. Overall, the proposed end-to-end approach promises significantly improved efficiency and performance for on-device health monitoring on readily available consumer electronics.

arXiv:2204.08624

Topology and geometry of data manifold in deep learning

Despite significant advances in the field of deep learning in applications to various fields, explaining the inner processes of deep learning models remains an important and open question. The purpose of this article is to describe and substantiate the geometric and topological view of the learning process of neural networks. Our attention is focused on the internal representation of neural networks and on the dynamics of changes in the topology and geometry of the data manifold on different layers. We also propose a method for assessing the generalizing ability of neural networks based on topological descriptors. In this paper, we use the concepts of topological data analysis and intrinsic dimension, and we present a wide range of experiments on different datasets and different configurations of convolutional neural network architectures. In addition, we consider the issue of the geometry of adversarial attacks in the classification task and spoofing attacks on face recognition systems. Our work is a contribution to the development of an important area of explainable and interpretable AI through the example of computer vision.