PyTorchKR Papers

최신 논문

총 12,350편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1812.06499공식 코드 · Papers with Code

HoVer-Net: Simultaneous Segmentation and Classification of Nuclei in Multi-Tissue Histology Images

Nuclear segmentation and classification within Haematoxylin & Eosin stained histology images is a fundamental prerequisite in the digital pathology work-flow. The development of automated methods for nuclear segmentation and classification enables the quantitative analysis of tens of thousands of nuclei within a whole-slide pathology image, opening up possibilities of further analysis of large-scale nuclear morphometry. However, automated nuclear segmentation and classification is faced with a major challenge in that there are several different types of nuclei, some of them exhibiting large intra-class variability such as the tumour cells. Additionally, some of the nuclei are often clustered together. To address these challenges, we present a novel convolutional neural network for simultaneous nuclear segmentation and classification that leverages the instance-rich information encoded within the vertical and horizontal distances of nuclear pixels to their centres of mass. These distances are then utilised to separate clustered nuclei, resulting in an accurate segmentation, particularly in areas with overlapping instances. Then for each segmented instance, the network predicts the type of nucleus via a devoted up-sampling branch. We demonstrate state-of-the-art performance compared to other methods on multiple independent multi-tissue histology image datasets. As part of this work, we introduce a new dataset of Haematoxylin & Eosin stained colorectal adenocarcinoma image tiles, containing 24,319 exhaustively annotated nuclei with associated class labels.

arXiv:1812.06408

Human Pose and Path Estimation from Aerial Video using Dynamic Classifier Selection

We consider the problem of estimating human pose and trajectory by an aerial robot with a monocular camera in near real time. We present a preliminary solution whose distinguishing feature is a dynamic classifier selection architecture. In our solution, each video frame is corrected for perspective using projective transformation. Then, two alternative feature sets are used: (i) Histogram of Oriented Gradients (HOG) of the silhouette, (ii) Convolutional Neural Network (CNN) features of the RGB image. The features (HOG or CNN) are classified using a dynamic classifier. A class is defined as a pose-viewpoint pair, and a total of 64 classes are defined to represent a forward walking and turning gait sequence. Our solution provides three main advantages: (i) Classification is efficient due to dynamic selection (4-class vs. 64-class classification). (ii) Classification errors are confined to neighbors of the true view-points. (iii) The robust temporal relationship between poses is used to resolve the left-right ambiguities of human silhouettes. Experiments conducted on both fronto-parallel videos and aerial videos confirm our solution can achieve accurate pose and trajectory estimation for both scenarios. We found using HOG features provides higher accuracy than using CNN features. For example, applying the HOG-based variant of our scheme to the 'walking on a figure 8-shaped path' dataset (1652 frames) achieved estimation accuracies of 99.6% for viewpoints and 96.2% for number of poses.

arXiv:1812.06391

Fast MVAE: Joint separation and classification of mixed sources based on multichannel variational autoencoder with auxiliary classifier

This paper proposes an alternative algorithm for multichannel variational autoencoder (MVAE), a recently proposed multichannel source separation approach. While MVAE is notable in its impressive source separation performance, the convergence-guaranteed optimization algorithm and that it allows us to estimate source-class labels simultaneously with source separation, there are still two major drawbacks, i.e., the high computational complexity and unsatisfactory source classification accuracy. To overcome these drawbacks, the proposed method employs an auxiliary classifier VAE, an information-theoretic extension of the conditional VAE, for learning the generative model of the source spectrograms. Furthermore, with the trained auxiliary classifier, we introduce a novel algorithm for the optimization that is able to not only reduce the computational time but also improve the source classification performance. We call the proposed method "fast MVAE (fMVAE)". Experimental evaluations revealed that fMVAE achieved comparative source separation performance to MVAE and about 80% source classification accuracy rate while it reduced about 93% computational time.

arXiv:1812.06387

Pre-Trained Convolutional Neural Network Features for Facial Expression Recognition

Facial expression recognition has been an active area in computer vision with application areas including animation, social robots, personalized banking, etc. In this study, we explore the problem of image classification for detecting facial expressions based on features extracted from pre-trained convolutional neural networks trained on ImageNet database. Features are extracted and transferred to a Linear Support Vector Machine for classification. All experiments are performed on two publicly available datasets such as JAFFE and CK+ database. The results show that representations learned from pre-trained networks for a task such as object recognition can be transferred, and used for facial expression recognition. Furthermore, for a small dataset, using features from earlier layers of the VGG19 network provides better classification accuracy. Accuracies of 92.26% and 92.86% were achieved for the CK+ and JAFFE datasets respectively.

arXiv:1812.06307

Generative adversarial networks for generation and classification of physical rehabilitation movement episodes

This article proposes a method for mathematical modeling of human movements related to patient exercise episodes performed during physical therapy sessions by using artificial neural networks. The generative adversarial network structure is adopted, whereby a discriminative and a generative model are trained concurrently in an adversarial manner. Different network architectures are examined, with the discriminative and generative models structured as deep subnetworks of hidden layers comprised of convolutional or recurrent computational units. The models are validated on a data set of human movements recorded with an optical motion tracker. The results demonstrate an ability of the networks for classification of new instances of motions, and for generation of motion examples that resemble the recorded motion sequences.

arXiv:1812.06292

A short review on Applications of Deep learning for Cyber security

Deep learning is an advanced model of traditional machine learning. This has the capability to extract optimal feature representation from raw input samples. This has been applied towards various use cases in cyber security such as intrusion detection, malware classification, android malware detection, spam and phishing detection and binary analysis. This paper outlines the survey of all the works related to deep learning based solutions for various cyber security use cases. Keywords: Deep learning, intrusion detection, malware detection, Android malware detection, spam & phishing detection, traffic analysis, binary analysis.

arXiv:1812.06254

3DTI-Net: Learn Inner Transform Invariant 3D Geometry Features using Dynamic GCN

Deep learning on point clouds has made a lot of progress recently. Many point cloud dedicated deep learning frameworks, such as PointNet and PointNet++, have shown advantages in accuracy and speed comparing to those using traditional 3D convolution algorithms. However, nearly all of these methods face a challenge, since the coordinates of the point cloud are decided by the coordinate system, they cannot handle the problem of 3D transform invariance properly. In this paper, we propose a general framework for point cloud learning. We achieve transform invariance by learning inner 3D geometry feature based on local graph representation, and propose a feature extraction network based on graph convolution network. Through experiments on classification and segmentation tasks, our method achieves state-of-the-art performance in rotated 3D object classification, and achieve competitive performance with the state-of-the-art in classification and segmentation tasks with fixed coordinate value.

arXiv:1812.06247

Flatten-T Swish: a thresholded ReLU-Swish-like activation function for deep learning

Activation functions are essential for deep learning methods to learn and perform complex tasks such as image classification. Rectified Linear Unit (ReLU) has been widely used and become the default activation function across the deep learning community since 2012. Although ReLU has been popular, however, the hard zero property of the ReLU has heavily hindered the negative values from propagating through the network. Consequently, the deep neural network has not been benefited from the negative representations. In this work, an activation function called Flatten-T Swish (FTS) that leverage the benefit of the negative values is proposed. To verify its performance, this study evaluates FTS with ReLU and several recent activation functions. Each activation function is trained using MNIST dataset on five different deep fully connected neural networks (DFNNs) with depth vary from five to eight layers. For a fair evaluation, all DFNNs are using the same configuration settings. Based on the experimental results, FTS with a threshold value, T=-0.20 has the best overall performance. As compared with ReLU, FTS (T=-0.20) improves MNIST classification accuracy by 0.13%, 0.70%, 0.67%, 1.07% and 1.15% on wider 5 layers, slimmer 5 layers, 6 layers, 7 layers and 8 layers DFNNs respectively. Apart from this, the study also noticed that FTS converges twice as fast as ReLU. Although there are other existing activation functions are also evaluated, this study elects ReLU as the baseline activation function.

arXiv:1812.06944

Domain Adaptation on Graphs by Learning Graph Topologies: Theoretical Analysis and an Algorithm

Traditional machine learning algorithms assume that the training and test data have the same distribution, while this assumption does not necessarily hold in real applications. Domain adaptation methods take into account the deviations in the data distribution. In this work, we study the problem of domain adaptation on graphs. We consider a source graph and a target graph constructed with samples drawn from data manifolds. We study the problem of estimating the unknown class labels on the target graph using the label information on the source graph and the similarity between the two graphs. We particularly focus on a setting where the target label function is learnt such that its spectrum is similar to that of the source label function. We first propose a theoretical analysis of domain adaptation on graphs and present performance bounds that characterize the target classification error in terms of the properties of the graphs and the data manifolds. We show that the classification performance improves as the topologies of the graphs get more balanced, i.e., as the numbers of neighbors of different graph nodes become more proportionate, and weak edges with small weights are avoided. Our results also suggest that graph edges between too distant data samples should be avoided for good generalization performance. We then propose a graph domain adaptation algorithm inspired by our theoretical findings, which estimates the label functions while learning the source and target graph topologies at the same time. The joint graph learning and label estimation problem is formulated through an objective function relying on our performance bounds, which is minimized with an alternating optimization scheme. Experiments on synthetic and real data sets suggest that the proposed method outperforms baseline approaches.

arXiv:1812.06858

Winter Road Surface Condition Recognition Using A Pretrained Deep Convolutional Network

This paper investigates the application of the latest machine learning technique deep neural networks for classifying road surface conditions (RSC) based on images from smartphones. Traditional machine learning techniques such as support vector machine (SVM) and random forests (RF) have been attempted in literature; however, their classification performance has been less than desirable due to challenges associated with image noises caused by sunlight glare and residual salts. A deep learning model based on convolutional neural network (CNN) is proposed and evaluated for its potential to address these challenges for improved classification accuracy. In the proposed approach we introduce the idea of applying an existing CNN model that has been pre-trained using millions of images with proven high recognition accuracy. The model is extended with two additional fully-connected layers of neurons for learning the specific features of the RSC images. The whole model is then trained with a low learning rate for fine-tuning by using a small set of RSC images. Results show that the proposed model has the highest classification performance in comparison to the traditional machine learning techniques. The testing accuracy with different training dataset sizes is also analyzed, showing the potential of achieving much higher accuracy with a larger training dataset.

arXiv:1812.05984

Measuring Similarity: Computationally Reproducing the Scholar's Interests

Computerized document classification already orders the news articles that Apple's "News" app or Google's "personalized search" feature groups together to match a reader's interests. The invisible and therefore illegible decisions that go into these tailored searches have been the subject of a critique by scholars who emphasize that our intelligence about documents is only as good as our ability to understand the criteria of search. This article will attempt to unpack the procedures used in computational classification of texts, translating them into term legible to humanists, and examining opportunities to render the computational text classification process subject to expert critique and improvement.

arXiv:1812.05980

Probabilistic Class-Specific Discriminant Analysis

In this paper we formulate a probabilistic model for class-specific discriminant subspace learning. The proposed model can naturally incorporate the multi-modal structure of the negative class, which is neglected by existing class-specific methods. Moreover, it can be directly used to define a class-specific probabilistic classification rule in the discriminant subspace. We show that existing class-specific discriminant analysis methods are special cases of the proposed probabilistic model and, by casting them as probabilistic models, they can be extended to class-specific classifiers. We illustrate the performance of the proposed model in both verification and classification problems.

arXiv:1812.05944

A Tutorial on Distance Metric Learning: Mathematical Foundations, Algorithms, Experimental Analysis, Prospects and Challenges (with Appendices on Mathematical Background and Detailed Algorithms Explanation)

Distance metric learning is a branch of machine learning that aims to learn distances from the data, which enhances the performance of similarity-based algorithms. This tutorial provides a theoretical background and foundations on this topic and a comprehensive experimental analysis of the most-known algorithms. We start by describing the distance metric learning problem and its main mathematical foundations, divided into three main blocks: convex analysis, matrix analysis and information theory. Then, we will describe a representative set of the most popular distance metric learning methods used in classification. All the algorithms studied in this paper will be evaluated with exhaustive testing in order to analyze their capabilities in standard classification problems, particularly considering dimensionality reduction and kernelization. The results, verified by Bayesian statistical tests, highlight a set of outstanding algorithms. Finally, we will discuss several potential future prospects and challenges in this field. This tutorial will serve as a starting point in the domain of distance metric learning from both a theoretical and practical perspective.

arXiv:1812.05836공식 코드 · Papers with Code

Rethinking Layer-wise Feature Amounts in Convolutional Neural Network Architectures

We characterize convolutional neural networks with respect to the relative amount of features per layer. Using a skew normal distribution as a parametrized framework, we investigate the common assumption of monotonously increasing feature-counts with higher layers of architecture designs. Our evaluation on models with VGG-type layers on the MNIST, Fashion-MNIST and CIFAR-10 image classification benchmarks provides evidence that motivates rethinking of our common assumption: architectures that favor larger early layers seem to yield better accuracy.

arXiv:1812.05792

Making Sense of Random Forest Probabilities: a Kernel Perspective

A random forest is a popular tool for estimating probabilities in machine learning classification tasks. However, the means by which this is accomplished is unprincipled: one simply counts the fraction of trees in a forest that vote for a certain class. In this paper, we forge a connection between random forests and kernel regression. This places random forest probability estimation on more sound statistical footing. As part of our investigation, we develop a model for the proximity kernel and relate it to the geometry and sparsity of the estimation problem. We also provide intuition and recommendations for tuning a random forest to improve its probability estimates.

arXiv:1812.05774

Don't Classify, Translate: Multi-Level E-Commerce Product Categorization Via Machine Translation

E-commerce platforms categorize their products into a multi-level taxonomy tree with thousands of leaf categories. Conventional methods for product categorization are typically based on machine learning classification algorithms. These algorithms take product information as input (e.g., titles and descriptions) to classify a product into a leaf category. In this paper, we propose a new paradigm based on machine translation. In our approach, we translate a product's natural language description into a sequence of tokens representing a root-to-leaf path in a product taxonomy. In our experiments on two large real-world datasets, we show that our approach achieves better predictive accuracy than a state-of-the-art classification system for product categorization. In addition, we demonstrate that our machine translation models can propose meaningful new paths between previously unconnected nodes in a taxonomy tree, thereby transforming the taxonomy into a directed acyclic graph (DAG). We discuss how the resultant taxonomy DAG promotes user-friendly navigation, and how it is more adaptable to new products.

arXiv:1812.05737

Effectiveness of Hierarchical Softmax in Large Scale Classification Tasks

Typically, Softmax is used in the final layer of a neural network to get a probability distribution for output classes. But the main problem with Softmax is that it is computationally expensive for large scale data sets with large number of possible outputs. To approximate class probability efficiently on such large scale data sets we can use Hierarchical Softmax. LSHTC datasets were used to study the performance of the Hierarchical Softmax. LSHTC datasets have large number of categories. In this paper we evaluate and report the performance of normal Softmax Vs Hierarchical Softmax on LSHTC datasets. This evaluation used macro f1 score as a performance measure. The observation was that the performance of Hierarchical Softmax degrades as the number of classes increase.

arXiv:1812.05720공식 코드 · Papers with Code

Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem

Classifiers used in the wild, in particular for safety-critical systems, should not only have good generalization properties but also should know when they don't know, in particular make low confidence predictions far away from the training data. We show that ReLU type neural networks which yield a piecewise linear classifier function fail in this regard as they produce almost always high confidence predictions far away from the training data. For bounded domains like images we propose a new robust optimization technique similar to adversarial training which enforces low confidence predictions far away from the training data. We show that this technique is surprisingly effective in reducing the confidence of predictions far away from the training data while maintaining high confidence predictions and test error on the original classification task compared to standard training.

arXiv:1812.05530

Combining Sentinel-1 and Sentinel-2 Time Series via RNN for object-based land cover classification

Radar and Optical Satellite Image Time Series (SITS) are sources of information that are commonly employed to monitor earth surfaces for tasks related to ecology, agriculture, mobility, land management planning and land cover monitoring. Many studies have been conducted using one of the two sources, but how to smartly combine the complementary information provided by radar and optical SITS is still an open challenge. In this context, we propose a new neural architecture for the combination of Sentinel-1 (S1) and Sentinel-2 (S2) imagery at object level, applied to a real-world land cover classification task. Experiments carried out on the Reunion Island, a overseas department of France in the Indian Ocean, demonstrate the significance of our proposal.

arXiv:1812.05473

Learning Features of Network Structures Using Graphlets

Networks are fundamental to the study of complex systems, ranging from social contacts, message transactions, to biological regulations and economical networks. In many realistic applications, these networks may vary over time. Modeling and analyzing such temporal properties is of additional interest as it can provide a richer characterization of relations between nodes in networks. In this paper, we explore the role of \emph{graphlets} in network classification for both static and temporal networks. Graphlets are small non-isomorphic induced subgraphs representing connected patterns in a network and their frequency can be used to assess network structures. We show that graphlet features, which are not captured by state-of-the-art methods, play a significant role in enhancing the performance of network classification. To that end, we propose two novel graphlet-based techniques, \emph{gl2vec} for network embedding, and \emph{gl-DCNN} for diffusion-convolutional neural networks. We demonstrate the efficacy and usability of \emph{gl2vec} and \emph{gl-DCNN} through extensive experiments using several real-world static and temporal networks. We find that features learned from graphlets can bring notable performance increases to state-of-the-art methods in network analysis.