Towards automated retinal screening, this paper makes an endeavor to simultaneously achieve pixel-level retinal lesion segmentation and image-level disease classification. Such a multi-task approach is crucial for accurate and clinically interpretable disease diagnosis. Prior art is insufficient due to three challenges, i.e., lesions lacking objective boundaries, clinical importance of lesions irrelevant to their size, and the lack of one-to-one correspondence between lesion and disease classes. This paper attacks the three challenges in the context of diabetic retinopathy (DR) grading. We propose Lesion-Net, a new variant of fully convolutional networks, with its expansive path re-designed to tackle the first challenge. A dual Dice loss that leverages both semantic segmentation and image classification losses is introduced to resolve the second challenge. Lastly, we build a multi-task network that employs Lesion-Net as a side-attention branch for both DR grading and result interpretation. A set of 12K fundus images is manually segmented by 45 ophthalmologists for 8 DR-related lesions, resulting in 290K manual segments in total. Extensive experiments on this large-scale dataset show that our proposed approach surpasses the prior art for multiple tasks including lesion segmentation, lesion classification and DR grading
In this paper, we present an InSphereNet method for the problem of 3D object classification. Unlike previous methods that use points, voxels, or multi-view images as inputs of deep neural network (DNN), the proposed method constructs a class of more representative features named infilling spheres from signed distance field (SDF). Because of the admirable spatial representation of infilling spheres, we can not only utilize very fewer number of spheres to accomplish classification task, but also design a lightweight InSphereNet with less layers and parameters than previous methods. Experiments on ModelNet40 show that the proposed method leads to superior performance than PointNet and PointNet++ in accuracy. In particular, if there are only a few dozen sphere inputs or about 100000 DNN parameters, the accuracy of our method remains at a very high level (over 88%). This further validates the conciseness and effectiveness of the proposed InSphere 3D representation. Keywords: 3D object classification , signed distance field , deep learning , infilling sphere
In order to enhance the performance of image recognition, a sparsity augmented probabilistic collaborative representation based classification (SA-ProCRC) method is presented. The proposed method obtains the dense coefficient through ProCRC, then augments the dense coefficient with a sparse one, and the sparse coefficient is attained by the orthogonal matching pursuit (OMP) algorithm. In contrast to conventional methods which require explicit computation of the reconstruction residuals for each class, the proposed method employs the augmented coefficient and the label matrix of the training samples to classify the test sample. Experimental results indicate that the proposed method can achieve promising results for face and scene images. The source code of our proposed SA-ProCRC is accessible at https://github.com/yinhefeng/SAProCRC.
One-class classification (OCC) deals with the classification problem in which the training data has data points belonging only to target class. In this paper, we study a one-class classification algorithm, One-Class Classification by Ensembles of Regression models (OCCER), that uses regression methods to address OCC problems. The OCCER coverts an OCC problem into many regression problems in the original feature space so that each feature of the original feature space is used as the target variable in one of the regression problems. Other features are used as the variables on which the dependent variable depends. The errors of regression of a data point by all the regression models are used to compute the outlier score of the data point. An extensive comparison of the OCCER algorithm with state-of-the-art OCC algorithms on several datasets was conducted to show the effectiveness of the this approach. We also demonstrate that the OCCER algorithm can work well with the latent feature space created by autoencoders for image datasets. The implementation of OCCER is available at https://github.com/srikanthBezawada/OCCER.
Deep learning methods have played a more and more important role in hyperspectral image classification. However, the general deep learning methods mainly take advantage of the information of sample itself or the pairwise information between samples while ignore the intrinsic data structure within the whole data. To tackle this problem, this work develops a novel deep manifold embedding method(DMEM) for hyperspectral image classification. First, each class in the image is modelled as a specific nonlinear manifold and the geodesic distance is used to measure the correlation between the samples. Then, based on the hierarchical clustering, the manifold structure of the data can be captured and each nonlinear data manifold can be divided into several sub-classes. Finally, considering the distribution of each sub-class and the correlation between different subclasses, the DMEM is constructed to preserve the estimated geodesic distances on the data manifold between the learned low dimensional features of different samples. Experiments over three real-world hyperspectral image datasets have demonstrated the effectiveness of the proposed method.
Deep learning has been used in the research of malware analysis. Most classification methods use either static analysis features or dynamic analysis features for malware family classification, and rarely combine them as classification features and also no extra effort is spent integrating the two types of features. In this paper, we combine static and dynamic analysis features with deep neural networks for Windows malware classification. We develop several methods to generate static and dynamic analysis features to classify malware in different ways. Given these features, we conduct experiments with composite neural network, showing that the proposed approach performs best with an accuracy of 83.17% on a total of 80 malware families with 4519 malware samples. Additionally, we show that using integrated features for malware family classification outperforms using static features or dynamic features alone. We show how static and dynamic features complement each other for malware classification.
In spite of the recent progresses on classifying 3D point cloud with deep CNNs, large geometric transformations like rotation and translation remain challenging problem and harm the final classification performance. To address this challenge, we propose Geometry Sharing Network (GS-Net) which effectively learns point descriptors with holistic context to enhance the robustness to geometric transformations. Compared with previous 3D point CNNs which perform convolution on nearby points, GS-Net can aggregate point features in a more global way. Specially, GS-Net consists of Geometry Similarity Connection (GSC) modules which exploit Eigen-Graph to group distant points with similar and relevant geometric information, and aggregate features from nearest neighbors in both Euclidean space and Eigenvalue space. This design allows GS-Net to efficiently capture both local and holistic geometric features such as symmetry, curvature, convexity and connectivity. Theoretically, we show the nearest neighbors of each point in Eigenvalue space are invariant to rotation and translation. We conduct extensive experiments on public datasets, ModelNet40, ShapeNet Part. Experiments demonstrate that GS-Net achieves the state-of-the-art performances on major datasets, 93.3% on ModelNet40, and are more robust to geometric transformations.
This paper presents a hybrid machine learning method of classifying residential requests in natural language to responsible departments that provide timely responses back to residents under the vision of digital government services in smart cities. Residential requests in natural language descriptions cover almost every aspect of a city's daily operation. Hence the responsible departments are fine-grained to even the level of local communities. There are no specific general categories or labels for each request sample. This causes two issues for supervised classification solutions, namely (1) the request sample data is unbalanced and (2) lack of specific labels for training. To solve these issues, we investigate a hybrid machine learning method that generates meta-class labels by means of unsupervised clustering algorithms; applies two-word embedding methods with three classifiers (including two hierarchical classifiers and one residual convolutional neural network); and selects the best performing classifier as the classification result. We demonstrate our approach performing better classification tasks compared to two benchmarking machine learning models, Naive Bayes classifier and a Multiple Layer Perceptron (MLP). In addition, the hierarchical classification method provides insights into the source of classification errors.
In this work, we conduct an extensive comparison of various approaches to speech based emotion recognition systems. The analyses were carried out on audio recordings from Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS). After pre-processing the raw audio files, features such as Log-Mel Spectrogram, Mel-Frequency Cepstral Coefficients (MFCCs), pitch and energy were considered. The significance of these features for emotion classification was compared by applying methods such as Long Short Term Memory (LSTM), Convolutional Neural Networks (CNNs), Hidden Markov Models (HMMs) and Deep Neural Networks (DNNs). On the 14-class (2 genders x 7 emotions) classification task, an accuracy of 68% was achieved with a 4-layer 2 dimensional CNN using the Log-Mel Spectrogram features. We also observe that, in emotion recognition, the choice of audio features impacts the results much more than the model complexity.
Recent semi-supervised learning methods have shown to achieve comparable results to their supervised counterparts while using only a small portion of labels in image classification tasks thanks to their regularization strategies. In this paper, we take a more direct approach for semi-supervised learning and propose learning to impute the labels of unlabeled samples such that a network achieves better generalization when it is trained on these labels. We pose the problem in a learning-to-learn formulation which can easily be incorporated to the state-of-the-art semi-supervised techniques and boost their performance especially when the labels are limited. We demonstrate that our method is applicable to both classification and regression problems including image classification and facial landmark detection tasks.
We survey various knowledge distillation (KD) strategies for simple classification tasks and implement a set of techniques that claim state-of-the-art accuracy. Our experiments using standardized model architectures, fixed compute budgets, and consistent training schedules indicate that many of these distillation results are hard to reproduce. This is especially apparent with methods using some form of feature distillation. Further examination reveals a lack of generalizability where these techniques may only succeed for specific architectures and training settings. We observe that appropriately tuned classical distillation in combination with a data augmentation training scheme gives an orthogonal improvement over other techniques. We validate this approach and open-source our code.
Learning knowledge representation is an increasingly important technology that supports a variety of machine learning related applications. However, the choice of hyperparameters is seldom justified and usually relies on exhaustive search. Understanding the effect of hyperparameter combinations on embedding quality is crucial to avoid the inefficient process and enhance practicality of vector representation methods. We evaluate the effects of distinct values for the margin parameter focused on translational embedding representation models for multi-relational categorized data. We assess the margin influence regarding the quality of embedding models by contrasting traditional link prediction task accuracy against a classification task. The findings provide evidence that lower values of margin are not rigorous enough to help with the learning process, whereas larger values produce much noise pushing the entities beyond to the surface of the hyperspace, thus requiring constant regularization. Finally, the correlation between link prediction and classification accuracy shows traditional validation protocol for embedding models is a weak metric to represent the quality of embedding representation.
Experimental reproducibility and replicability are critical topics in machine learning. Authors have often raised concerns about their lack in scientific publications to improve the quality of the field. Recently, the graph representation learning field has attracted the attention of a wide research community, which resulted in a large stream of works. As such, several Graph Neural Network models have been developed to effectively tackle graph classification. However, experimental procedures often lack rigorousness and are hardly reproducible. Motivated by this, we provide an overview of common practices that should be avoided to fairly compare with the state of the art. To counter this troubling trend, we ran more than 47000 experiments in a controlled and uniform framework to re-evaluate five popular models across nine common benchmarks. Moreover, by comparing GNNs with structure-agnostic baselines we provide convincing evidence that, on some datasets, structural information has not been exploited yet. We believe that this work can contribute to the development of the graph learning field, by providing a much needed grounding for rigorous evaluations of graph classification models.
Semmelhack et al. (2014) have achieved high classification accuracy in distinguishing swim bouts of zebrafish using a Support Vector Machine (SVM). Convolutional Neural Networks (CNNs) have reached superior performance in various image recognition tasks over SVMs, but these powerful networks remain a black box. Reaching better transparency helps to build trust in their classifications and makes learned features interpretable to experts. Using a recently developed technique called Deep Taylor Decomposition, we generated heatmaps to highlight input regions of high relevance for predictions. We find that our CNN makes predictions by analyzing the steadiness of the tail's trunk, which markedly differs from the manually extracted features used by Semmelhack et al. (2014). We further uncovered that the network paid attention to experimental artifacts. Removing these artifacts ensured the validity of predictions. After correction, our best CNN beats the SVM by 6.12%, achieving a classification accuracy of 96.32%. Our work thus demonstrates the utility of AI explainability for CNNs.
We propose a unified game-theoretical framework to perform classification and conditional image generation given limited supervision. It is formulated as a three-player minimax game consisting of a generator, a classifier and a discriminator, and therefore is referred to as Triple Generative Adversarial Network (Triple-GAN). The generator and the classifier characterize the conditional distributions between images and labels to perform conditional generation and classification, respectively. The discriminator solely focuses on identifying fake image-label pairs. Under a nonparametric assumption, we prove the unique equilibrium of the game is that the distributions characterized by the generator and the classifier converge to the data distribution. As a byproduct of the three-player mechanism, Triple-GAN is flexible to incorporate different semi-supervised classifiers and GAN architectures. We evaluate Triple-GAN in two challenging settings, namely, semi-supervised learning and the extreme low data regime. In both settings, Triple-GAN can achieve excellent classification results and generate meaningful samples in a specific class simultaneously. In particular, using a commonly adopted 13-layer CNN classifier, Triple-GAN outperforms extensive semi-supervised learning methods substantially on more than 10 benchmarks no matter data augmentation is applied or not.
A recently introduced text classifier, called SS3, has obtained state-of-the-art performance on the CLEF's eRisk tasks. SS3 was created to deal with risk detection over text streams and, therefore, not only supports incremental training and classification but also can visually explain its rationale. However, little attention has been paid to the potential use of SS3 as a general classifier. We believe this could be due to the unavailability of an open-source implementation of SS3. In this work, we introduce PySS3, a package that implements SS3 and also comes with visualization tools that allow researchers to deploy robust, explainable, and trusty machine learning models for text classification.
In this paper we address three different aspects of semantic segmentation from remote sensor data using deep neural networks. Firstly, we focus on the semantic segmentation of buildings from remote sensor data and propose ICT-Net. The proposed network has been tested on the INRIA and AIRS benchmark datasets and is shown to outperform all other state of the art by more than 1.5% and 1.8% on the Jaccard index, respectively. Secondly, as the building classification is typically the first step of the reconstruction process, we investigate the relationship of the classification accuracy to the reconstruction accuracy. Finally, we present the simple yet compelling concept of latent learning and the implications it carries within the context of deep learning. We posit that a network trained on a primary task (i.e. building classification) is unintentionally learning about auxiliary tasks (e.g. the classification of road, tree, etc) which are complementary to the primary task. We extensively tested the proposed technique on the ISPRS benchmark dataset which contains multi-label ground truth, and report an average classification accuracy (F1 score) of 54.29% (SD=17.03) for roads, 10.15% (SD=2.54) for cars, 24.11% (SD=5.25) for trees, 42.74% (SD=6.62) for low vegetation, and 18.30% (SD=16.08) for clutter. The source code and supplemental material is publicly available at http://www.theICTlab.org/lp/2019ICT-Net/.
The ubiquitous availability of wearable sensors is responsible for driving the Internet-of-Things but is also making an impact on sport sciences and precision medicine. While human activity recognition from smartphone data or other types of inertial measurement units (IMU) has evolved to one of the most prominent daily life examples of machine learning, the underlying process of time-series feature engineering still seems to be time-consuming. This lengthy process inhibits the development of IMU-based machine learning applications in sport science and precision medicine. This contribution discusses a feature engineering workflow, which automates the extraction of time-series feature on based on the FRESH algorithm (FeatuRe Extraction based on Scalable Hypothesis tests) to identify statistically significant features from synchronized IMU sensors (IMeasureU Ltd, NZ). The feature engineering workflow has five main steps: time-series engineering, automated time-series feature extraction, optimized feature extraction, fitting of a specialized classifier, and deployment of optimized machine learning pipeline. The workflow is discussed for the case of a user-specific running-walking classification, and the generalization to a multi-user multi-activity classification is demonstrated.
We propose a new supervised learning algorithm, for classification and regression problems where two or more preliminary predictors are available. We introduce \texttt{KernelCobra}, a non-linear learning strategy for combining an arbitrary number of initial predictors. \texttt{KernelCobra} builds on the COBRA algorithm introduced by \citet{biau2016cobra}, which combined estimators based on a notion of proximity of predictions on the training data. While the COBRA algorithm used a binary threshold to declare which training data were close and to be used, we generalize this idea by using a kernel to better encapsulate the proximity information. Such a smoothing kernel provides more representative weights to each of the training points which are used to build the aggregate and final predictor, and \texttt{KernelCobra} systematically outperforms the COBRA algorithm. While COBRA is intended for regression, \texttt{KernelCobra} deals with classification and regression. \texttt{KernelCobra} is included as part of the open source Python package \texttt{Pycobra} (0.2.4 and onward), introduced by \citet{guedj2018pycobra}. Numerical experiments assess the performance (in terms of pure prediction and computational complexity) of \texttt{KernelCobra} on real-life and synthetic datasets.
Kernel methods have had great success in Statistics and Machine Learning. Despite their growing popularity, however, less effort has been drawn towards developing kernel based classification methods on Riemannian manifolds due to difficulty in dealing with non-Euclidean geometry. In this paper, motivated by the extrinsic framework of manifold-valued data analysis, we propose a new positive definite kernel on planar Kendall shape space $\Sigma_2^k$, called extrinsic Veronese Whitney Gaussian kernel. We show that our approach can be extended to develop Gaussian kernels on any embedded manifold. Furthermore, kernel ridge regression classifier (KRRC) is implemented to address the shape classification problem on $\Sigma_2^k$, and their promising performances are illustrated through the real data analysis.