PyTorchKR Papers

최신 논문

총 12,350편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1402.1947

Classification Tree Diagrams in Health Informatics Applications

Health informatics deal with the methods used to optimize the acquisition, storage and retrieval of medical data, and classify information in healthcare applications. Healthcare analysts are particularly interested in various computer informatics areas such as; knowledge representation from data, anomaly detection, outbreak detection methods and syndromic surveillance applications. Although various parametric and non-parametric approaches are being proposed to classify information from data, classification tree diagrams provide an interactive visualization to analysts as compared to other methods. In this work we discuss application of classification tree diagrams to classify information from medical data in healthcare applications.

arXiv:1402.1371

Quantile Representation for Indirect Immunofluorescence Image Classification

In the diagnosis of autoimmune diseases, an important task is to classify images of slides containing several HEp-2 cells. All cells from one slide share the same label, and by classifying cells from one slide independently, some information on the global image quality and intensity is lost. Considering one whole slide as a collection (a bag) of feature vectors, however, poses the problem of how to handle this bag. A simple, and surprisingly effective, approach is to summarize the bag of feature vectors by a few quantile values per feature. This characterizes the full distribution of all instances, thereby assuming that all instances in a bag are informative. This representation is particularly useful when each bag contains many feature vectors, which is the case in the classification of the immunofluorescence images. Experiments on the classification of indirect immunofluorescence images show the usefulness of this approach.

arXiv:1312.6594

Sequentially Generated Instance-Dependent Image Representations for Classification

In this paper, we investigate a new framework for image classification that adaptively generates spatial representations. Our strategy is based on a sequential process that learns to explore the different regions of any image in order to infer its category. In particular, the choice of regions is specific to each image, directed by the actual content of previously selected regions.The capacity of the system to handle incomplete image information as well as its adaptive region selection allow the system to perform well in budgeted classification tasks by exploiting a dynamicly generated representation of each image. We demonstrate the system's abilities in a series of image-based exploration and classification tasks that highlight its learned exploration and inference abilities.

arXiv:1312.5604

Learning Transformations for Classification Forests

This work introduces a transformation-based learner model for classification forests. The weak learner at each split node plays a crucial role in a classification tree. We propose to optimize the splitting objective by learning a linear transformation on subspaces using nuclear norm as the optimization criteria. The learned linear transformation restores a low-rank structure for data from the same class, and, at the same time, maximizes the separation between different classes, thereby improving the performance of the split function. Theoretical and experimental results support the proposed framework.

arXiv:1309.7750

An Extensive Experimental Study on the Cluster-based Reference Set Reduction for speeding-up the k-NN Classifier

The k-Nearest Neighbor (k-NN) classification algorithm is one of the most widely-used lazy classifiers because of its simplicity and ease of implementation. It is considered to be an effective classifier and has many applications. However, its major drawback is that when sequential search is used to find the neighbors, it involves high computational cost. Speeding-up k-NN search is still an active research field. Hwang and Cho have recently proposed an adaptive cluster-based method for fast Nearest Neighbor searching. The effectiveness of this method is based on the adjustment of three parameters. However, the authors evaluated their method by setting specific parameter values and using only one dataset. In this paper, an extensive experimental study of this method is presented. The results, which are based on five real life datasets, illustrate that if the parameters of the method are carefully defined, one can achieve even better classification performance.

arXiv:1405.6177

Automated Fabric Defect Inspection: A Survey of Classifiers

Quality control at each stage of production in textile industry has become a key factor to retaining the existence in the highly competitive global market. Problems of manual fabric defect inspection are lack of accuracy and high time consumption, where early and accurate fabric defect detection is a significant phase of quality control. Computer vision based, i.e. automated fabric defect inspection systems are thought by many researchers of different countries to be very useful to resolve these problems. There are two major challenges to be resolved to attain a successful automated fabric defect inspection system. They are defect detection and defect classification. In this work, we discuss different techniques used for automated fabric defect classification, then show a survey of classifiers used in automated fabric defect inspection systems, and finally, compare these classifiers by using performance metrics. This work is expected to be very useful for the researchers in the area of automated fabric defect inspection to understand and evaluate the many potential options in this field.

arXiv:1402.4539

A Statistical Approach to Set Classification by Feature Selection with Applications to Classification of Histopathology Images

Set classification problems arise when classification tasks are based on sets of observations as opposed to individual observations. In set classification, a classification rule is trained with $N$ sets of observations, where each set is labeled with class information, and the prediction of a class label is performed also with a set of observations. Data sets for set classification appear, for example, in diagnostics of disease based on multiple cell nucleus images from a single tissue. Relevant statistical models for set classification are introduced, which motivate a set classification framework based on context-free feature extraction. By understanding a set of observations as an empirical distribution, we employ a data-driven method to choose those features which contain information on location and major variation. In particular, the method of principal component analysis is used to extract the features of major variation. Multidimensional scaling is used to represent features as vector-valued points on which conventional classifiers can be applied. The proposed set classification approaches achieve better classification results than competing methods in a number of simulated data examples. The benefits of our method are demonstrated in an analysis of histopathology images of cell nuclei related to liver cancer.

arXiv:1402.4380

A Comparative Study of Machine Learning Methods for Verbal Autopsy Text Classification

A Verbal Autopsy is the record of an interview about the circumstances of an uncertified death. In developing countries, if a death occurs away from health facilities, a field-worker interviews a relative of the deceased about the circumstances of the death; this Verbal Autopsy can be reviewed off-site. We report on a comparative study of the processes involved in Text Classification applied to classifying Cause of Death: feature value representation; machine learning classification algorithms; and feature reduction strategies in order to identify the suitable approaches applicable to the classification of Verbal Autopsy text. We demonstrate that normalised term frequency and the standard TFiDF achieve comparable performance across a number of classifiers. The results also show Support Vector Machine is superior to other classification algorithms employed in this research. Finally, we demonstrate the effectiveness of employing a "locally-semi-supervised" feature reduction strategy in order to increase performance accuracy.

arXiv:1402.3891

Performance Evaluation of Machine Learning Classifiers in Sentiment Mining

In recent years, the use of machine learning classifiers is of great value in solving a variety of problems in text classification. Sentiment mining is a kind of text classification in which, messages are classified according to sentiment orientation such as positive or negative. This paper extends the idea of evaluating the performance of various classifiers to show their effectiveness in sentiment mining of online product reviews. The product reviews are collected from Amazon reviews. To evaluate the performance of classifiers various evaluation methods like random sampling, linear sampling and bootstrap sampling are used. Our results shows that support vector machine with bootstrap sampling method outperforms others classifiers and sampling methods in terms of misclassification rate.

arXiv:1402.3511

A Clockwork RNN

Sequence prediction and classification are ubiquitous and challenging problems in machine learning that can require identifying complex dependencies between temporally distant inputs. Recurrent Neural Networks (RNNs) have the ability, in theory, to cope with these temporal dependencies by virtue of the short-term memory implemented by their recurrent (feedback) connections. However, in practice they are difficult to train successfully when the long-term memory is required. This paper introduces a simple, yet powerful modification to the standard RNN architecture, the Clockwork RNN (CW-RNN), in which the hidden layer is partitioned into separate modules, each processing inputs at its own temporal granularity, making computations only at its prescribed clock rate. Rather than making the standard RNN models more complex, CW-RNN reduces the number of RNN parameters, improves the performance significantly in the tasks tested, and speeds up the network evaluation. The network is demonstrated in preliminary experiments involving two tasks: audio signal generation and TIMIT spoken word classification, where it outperforms both RNN and LSTM networks.

arXiv:1312.6116

Improving Deep Neural Networks with Probabilistic Maxout Units

We present a probabilistic variant of the recently introduced maxout unit. The success of deep neural networks utilizing maxout can partly be attributed to favorable performance under dropout, when compared to rectified linear units. It however also depends on the fact that each maxout unit performs a pooling operation over a group of linear transformations and is thus partially invariant to changes in its input. Starting from this observation we ask the question: Can the desirable properties of maxout units be preserved while improving their invariance properties ? We argue that our probabilistic maxout (probout) units successfully achieve this balance. We quantitatively verify this claim and report classification performance matching or exceeding the current state of the art on three challenging image classification benchmarks (CIFAR-10, CIFAR-100 and SVHN).

arXiv:1312.0365

The Law of Total Odds

The law of total probability may be deployed in binary classification exercises to estimate the unconditional class probabilities if the class proportions in the training set are not representative of the population class proportions. We argue that this is not a conceptually sound approach and suggest an alternative based on the new law of total odds. We quantify the bias of the total probability estimator of the unconditional class probabilities and show that the total odds estimator is unbiased. The sample version of the total odds estimator is shown to coincide with a maximum-likelihood estimator known from the literature. The law of total odds can also be used for transforming the conditional class probabilities if independent estimates of the unconditional class probabilities of the population are available. Keywords: Total probability, likelihood ratio, Bayes' formula, binary classification, relative odds, unbiased estimator, supervised learning, dataset shift.

arXiv:1402.6859

Outlier Detection using Improved Genetic K-means

The outlier detection problem in some cases is similar to the classification problem. For example, the main concern of clustering-based outlier detection algorithms is to find clusters and outliers, which are often regarded as noise that should be removed in order to make more reliable clustering. In this article, we present an algorithm that provides outlier detection and data clustering simultaneously. The algorithmimprovesthe estimation of centroids of the generative distribution during the process of clustering and outlier discovery. The proposed algorithm consists of two stages. The first stage consists of improved genetic k-means algorithm (IGK) process, while the second stage iteratively removes the vectors which are far from their cluster centroids.

arXiv:1402.6383

Large-margin Learning of Compact Binary Image Encodings

The use of high-dimensional features has become a normal practice in many computer vision applications. The large dimension of these features is a limiting factor upon the number of data points which may be effectively stored and processed, however. We address this problem by developing a novel approach to learning a compact binary encoding, which exploits both pair-wise proximity and class-label information on training data set. Exploiting this extra information allows the development of encodings which, although compact, outperform the original high-dimensional features in terms of final classification or retrieval performance. The method is general, in that it is applicable to both non-parametric and parametric learning methods. This generality means that the embedded features are suitable for a wide variety of computer vision tasks, such as image classification and content-based image retrieval. Experimental results demonstrate that the new compact descriptor achieves an accuracy comparable to, and in some cases better than, the visual descriptor in the original space despite being significantly more compact. Moreover, any convex loss function and convex regularization penalty (e.g., $ \ell_p $ norm with $ p \ge 1 $) can be incorporated into the framework, which provides future flexibility.

arXiv:1402.5728

Machine Learning Methods in the Computational Biology of Cancer

The objectives of this "perspective" paper are to review some recent advances in sparse feature selection for regression and classification, as well as compressed sensing, and to discuss how these might be used to develop tools to advance personalized cancer therapy. As an illustration of the possibilities, a new algorithm for sparse regression is presented, and is applied to predict the time to tumor recurrence in ovarian cancer. A new algorithm for sparse feature selection in classification problems is presented, and its validation in endometrial cancer is briefly discussed. Some open problems are also presented.

arXiv:1402.4861

A Quasi-Newton Method for Large Scale Support Vector Machines

This paper adapts a recently developed regularized stochastic version of the Broyden, Fletcher, Goldfarb, and Shanno (BFGS) quasi-Newton method for the solution of support vector machine classification problems. The proposed method is shown to converge almost surely to the optimal classifier at a rate that is linear in expectation. Numerical results show that the proposed method exhibits a convergence rate that degrades smoothly with the dimensionality of the feature vectors.

arXiv:1312.6229

OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks

We present an integrated framework for using Convolutional Networks for classification, localization and detection. We show how a multiscale and sliding window approach can be efficiently implemented within a ConvNet. We also introduce a novel deep learning approach to localization by learning to predict object boundaries. Bounding boxes are then accumulated rather than suppressed in order to increase detection confidence. We show that different tasks can be learned simultaneously using a single shared network. This integrated framework is the winner of the localization task of the ImageNet Large Scale Visual Recognition Challenge 2013 (ILSVRC2013) and obtained very competitive results for the detection and classifications tasks. In post-competition work, we establish a new state of the art for the detection task. Finally, we release a feature extractor from our best model called OverFeat.

arXiv:1311.1294

Delay Learning Architectures for Memory and Classification

We present a neuromorphic spiking neural network, the DELTRON, that can remember and store patterns by changing the delays of every connection as opposed to modifying the weights. The advantage of this architecture over traditional weight based ones is simpler hardware implementation without multipliers or digital-analog converters (DACs) as well as being suited to time-based computing. The name is derived due to similarity in the learning rule with an earlier architecture called Tempotron. The DELTRON can remember more patterns than other delay-based networks by modifying a few delays to remember the most 'salient' or synchronous part of every spike pattern. We present simulations of memory capacity and classification ability of the DELTRON for different random spatio-temporal spike patterns. The memory capacity for noisy spike patterns and missing spikes are also shown. Finally, we present SPICE simulation results of the core circuits involved in a reconfigurable mixed signal implementation of this architecture.

arXiv:1211.3046

Recovering the Optimal Solution by Dual Random Projection

Random projection has been widely used in data classification. It maps high-dimensional data into a low-dimensional subspace in order to reduce the computational cost in solving the related optimization problem. While previous studies are focused on analyzing the classification performance of using random projection, in this work, we consider the recovery problem, i.e., how to accurately recover the optimal solution to the original optimization problem in the high-dimensional space based on the solution learned from the subspace spanned by random projections. We present a simple algorithm, termed Dual Random Projection, that uses the dual solution of the low-dimensional optimization problem to recover the optimal solution to the original problem. Our theoretical analysis shows that with a high probability, the proposed algorithm is able to accurately recover the optimal solution to the original problem, provided that the data matrix is of low rank or can be well approximated by a low rank matrix.