Clearly explaining a rationale for a classification decision to an end-user can be as important as the decision itself. Existing approaches for deep visual recognition are generally opaque and do not output any justification text; contemporary vision-language models can describe image content but fail to take into account class-discriminative image aspects which justify visual predictions. We propose a new model that focuses on the discriminating properties of the visible object, jointly predicts a class label, and explains why the predicted label is appropriate for the image. We propose a novel loss function based on sampling and reinforcement learning that learns to generate sentences that realize a global sentence property, such as class specificity. Our results on a fine-grained bird species classification dataset show that our model is able to generate explanations which are not only consistent with an image but also more discriminative than descriptions produced by existing captioning methods.
Sparseness is a useful regularizer for learning in a wide range of applications, in particular in neural networks. This paper proposes a model targeted at classification tasks, where sparse activity and sparse connectivity are used to enhance classification capabilities. The tool for achieving this is a sparseness-enforcing projection operator which finds the closest vector with a pre-defined sparseness for any given vector. In the theoretical part of this paper, a comprehensive theory for such a projection is developed. In conclusion, it is shown that the projection is differentiable almost everywhere and can thus be implemented as a smooth neuronal transfer function. The entire model can hence be tuned end-to-end using gradient-based methods. Experiments on the MNIST database of handwritten digits show that classification performance can be boosted by sparse activity or sparse connectivity. With a combination of both, performance can be significantly better compared to classical non-sparse approaches.
This paper focuses on two key problems for audio-visual emotion recognition in the video. One is the audio and visual streams temporal alignment for feature level fusion. The other one is locating and re-weighting the perception attentions in the whole audio-visual stream for better recognition. The Long Short Term Memory Recurrent Neural Network (LSTM-RNN) is employed as the main classification architecture. Firstly, soft attention mechanism aligns the audio and visual streams. Secondly, seven emotion embedding vectors, which are corresponding to each classification emotion type, are added to locate the perception attentions. The locating and re-weighting process is also based on the soft attention mechanism. The experiment results on EmotiW2015 dataset and the qualitative analysis show the efficiency of the proposed two techniques.
The goal of domain adaptation is to adapt models learned on a source domain to a particular target domain. Most methods for unsupervised domain adaptation proposed in the literature to date, assume that the set of classes present in the target domain is identical to the set of classes present in the source domain. This is a restrictive assumption that limits the practical applicability of unsupervised domain adaptation techniques in real world settings ("in the wild"). Therefore, we relax this constraint and propose a technique that allows the set of target classes to be a subset of the source classes. This way, large publicly available annotated datasets with a wide variety of classes can be used as source, even if the actual set of classes in target can be more limited and, maybe most importantly, unknown beforehand. To this end, we propose an algorithm that orders a set of source subspaces that are relevant to the target classification problem. Our method then chooses a restricted set from this ordered set of source subspaces. As an extension, even starting from multiple source datasets with varied sets of categories, this method automatically selects an appropriate subset of source categories relevant to a target dataset. Empirical analysis on a number of source and target domain datasets shows that restricting the source subspace to only a subset of categories does indeed substantially improve the eventual target classification accuracy over the baseline that considers all source classes.
Large medical image data sets with high dimensionality require substantial amount of computation time for data creation and data processing. This paper presents a novel generalized method that finds optimal image-based feature sets that reduce computational time complexity while maximizing overall classification accuracy for detection of diabetic retinopathy (DR). First, region-based and pixel-based features are extracted from fundus images for classification of DR lesions and vessel-like structures. Next, feature ranking strategies are used to distinguish the optimal classification feature sets. DR lesion and vessel classification accuracies are computed using the boosted decision tree and decision forest classifiers in the Microsoft Azure Machine Learning Studio platform, respectively. For images from the DIARETDB1 data set, 40 of its highest-ranked features are used to classify four DR lesion types with an average classification accuracy of 90.1% in 792 seconds. Also, for classification of red lesion regions and hemorrhages from microaneurysms, accuracies of 85% and 72% are observed, respectively. For images from STARE data set, 40 high-ranked features can classify minor blood vessels with an accuracy of 83.5% in 326 seconds. Such cloud-based fundus image analysis systems can significantly enhance the borderline classification performances in automated screening systems.
The constant growth in the present day real-world databases pose computational challenges for a single computer. Cloud-based platforms, on the other hand, are capable of handling large volumes of information manipulation tasks, thereby necessitating their use for large real-world data set computations. This work focuses on creating a novel Generalized Flow within the cloud-based computing platform: Microsoft Azure Machine Learning Studio (MAMLS) that accepts multi-class and binary classification data sets alike and processes them to maximize the overall classification accuracy. First, each data set is split into training and testing data sets, respectively. Then, linear and nonlinear classification model parameters are estimated using the training data set. Data dimensionality reduction is then performed to maximize classification accuracy. For multi-class data sets, data centric information is used to further improve overall classification accuracy by reducing the multi-class classification to a series of hierarchical binary classification tasks. Finally, the performance of optimized classification model thus achieved is evaluated and scored on the testing data set. The classification characteristics of the proposed flow are comparatively evaluated on 3 public data sets and a local data set with respect to existing state-of-the-art methods. On the 3 public data sets, the proposed flow achieves 78-97.5% classification accuracy. Also, the local data set, created using the information regarding presence of Diabetic Retinopathy lesions in fundus images, results in 85.3-95.7% average classification accuracy, which is higher than the existing methods. Thus, the proposed generalized flow can be useful for a wide range of application-oriented "big data sets".
We introduce a novel, simple convolution neural network (CNN) architecture - multi-group norm constraint CNN (MGNC-CNN) that capitalizes on multiple sets of word embeddings for sentence classification. MGNC-CNN extracts features from input embedding sets independently and then joins these at the penultimate layer in the network to form a final feature vector. We then adopt a group regularization strategy that differentially penalizes weights associated with the subcomponents generated from the respective embedding sets. This model is much simpler than comparable alternative architectures and requires substantially less training time. Furthermore, it is flexible in that it does not require input word embeddings to be of the same dimensionality. We show that MGNC-CNN consistently outperforms baseline models.
Object localization is an important computer vision problem with a variety of applications. The lack of large scale object-level annotations and the relative abundance of image-level labels makes a compelling case for weak supervision in the object localization task. Deep Convolutional Neural Networks are a class of state-of-the-art methods for the related problem of object recognition. In this paper, we describe a novel object localization algorithm which uses classification networks trained on only image labels. This weakly supervised method leverages local spatial and semantic patterns captured in the convolutional layers of classification networks. We propose an efficient beam search based approach to detect and localize multiple objects in images. The proposed method significantly outperforms the state-of-the-art in standard object localization data-sets with a 8 point increase in mAP scores.
Robust classification becomes challenging when each class consists of multiple subclasses. Examples include multi-font optical character recognition and automated protein function prediction. In correlation-based nearest-neighbor classification, the maximin correlation approach (MCA) provides the worst-case optimal solution by minimizing the maximum misclassification risk through an iterative procedure. Despite the optimality, the original MCA has drawbacks that have limited its wide applicability in practice. That is, the MCA tends to be sensitive to outliers, cannot effectively handle nonlinearities in datasets, and suffers from having high computational complexity. To address these limitations, we propose an improved solution, named regularized maximin correlation approach (R-MCA). We first reformulate MCA as a quadratically constrained linear programming (QCLP) problem, incorporate regularization by introducing slack variables in the primal problem of the QCLP, and derive the corresponding Lagrangian dual. The dual formulation enables us to apply the kernel trick to R-MCA so that it can better handle nonlinearities. Our experimental results demonstrate that the regularization and kernelization make the proposed R-MCA more robust and accurate for various classification tasks than the original MCA. Furthermore, when the data size or dimensionality grows, R-MCA runs substantially faster by solving either the primal or dual (whichever has a smaller variable dimension) of the QCLP.
The goal of this paper is to analyze an intriguing phenomenon recently discovered in deep networks, namely their instability to adversarial perturbations (Szegedy et. al., 2014). We provide a theoretical framework for analyzing the robustness of classifiers to adversarial perturbations, and show fundamental upper bounds on the robustness of classifiers. Specifically, we establish a general upper bound on the robustness of classifiers to adversarial perturbations, and then illustrate the obtained upper bound on the families of linear and quadratic classifiers. In both cases, our upper bound depends on a distinguishability measure that captures the notion of difficulty of the classification task. Our results for both classes imply that in tasks involving small distinguishability, no classifier in the considered set will be robust to adversarial perturbations, even if a good accuracy is achieved. Our theoretical framework moreover suggests that the phenomenon of adversarial instability is due to the low flexibility of classifiers, compared to the difficulty of the classification task (captured by the distinguishability). Moreover, we show the existence of a clear distinction between the robustness of a classifier to random noise and its robustness to adversarial perturbations. Specifically, the former is shown to be larger than the latter by a factor that is proportional to \sqrt{d} (with d being the signal dimension) for linear classifiers. This result gives a theoretical explanation for the discrepancy between the two robustness properties in high dimensional problems, which was empirically observed in the context of neural networks. To the best of our knowledge, our results provide the first theoretical work that addresses the phenomenon of adversarial instability recently observed for deep networks. Our analysis is complemented by experimental results on controlled and real-world data.
The growth in electronic transactions and human machine interactions rely on the information such as gender, age, expression and ethnicity provided by the face image. In order to obtain these information, feature extraction plays a major role. In this paper, retrieval of age, gender, expression and race information from an individual face image is analysed using different feature extraction methods. The performance of four major feature extraction methods such as Active Appearance Model (AAM), Gabor wavelets, Local Binary Pattern (LBP) and Wavelet Decomposition (WD) are analyzed for gender recognition, age estimation, expression recognition and racial recognition in terms of accuracy (recognition rate), time for feature extraction, neural training and time to test an image. Each of this recognition system is compared with four feature extractors on same dataset (training and validation set) to get a better understanding in its performance. Experiments carried out on FG-NET, Cohn-Kanade, PAL face database shows that each method has its own merits and demerits. Hence it is practically impossible to define a method which is best at all circumstances with less computational complexity. Further, a detailed comparison of age estimation and age estimation using gender information is provided along with a solution to overcome aging effect in case of gender recognition. An attempt has been made in obtaining all (i.e. gender, age range, expression and ethnicity) information from a test image in a single go.
The objective of this paper is the effective transfer of the Convolutional Neural Network (CNN) feature in image search and classification. Systematically, we study three facts in CNN transfer. 1) We demonstrate the advantage of using images with a properly large size as input to CNN instead of the conventionally resized one. 2) We benchmark the performance of different CNN layers improved by average/max pooling on the feature maps. Our observation suggests that the Conv5 feature yields very competitive accuracy under such pooling step. 3) We find that the simple combination of pooled features extracted across various CNN layers is effective in collecting evidences from both low and high level descriptors. Following these good practices, we are capable of improving the state of the art on a number of benchmarks to a large margin.
Recurrent neural network architectures combining with attention mechanism, or neural attention model, have shown promising performance recently for the tasks including speech recognition, image caption generation, visual question answering and machine translation. In this paper, neural attention model is applied on two sequence classification tasks, dialogue act detection and key term extraction. In the sequence labeling tasks, the model input is a sequence, and the output is the label of the input sequence. The major difficulty of sequence labeling is that when the input sequence is long, it can include many noisy or irrelevant part. If the information in the whole sequence is treated equally, the noisy or irrelevant part may degrade the classification performance. The attention mechanism is helpful for sequence classification task because it is capable of highlighting important part among the entire sequence for the classification task. The experimental results show that with the attention mechanism, discernible improvements were achieved in the sequence labeling task considered here. The roles of the attention mechanism in the tasks are further analyzed and visualized in this paper.
The workflow of extracting features from images using convolutional neural networks (CNN) and generating captions with recurrent neural networks (RNN) has become a de-facto standard for image captioning task. However, since CNN features are originally designed for classification task, it is mostly concerned with the main conspicuous element of the image, and often fails to correctly convey information on local, secondary elements. We propose to incorporate coding with vector of locally aggregated descriptors (VLAD) on spatial pyramid for CNN features of sub-regions in order to generate image representations that better reflect the local information of the images. Our results show that our method of compact VLAD coding can match CNN features with as little as 3% of dimensionality and, when combined with spatial pyramid, it results in image captions that more accurately take local elements into account.
In histopathological image analysis, feature extraction for classification is a challenging task due to the diversity of histology features suitable for each problem as well as presence of rich geometrical structures. In this paper, we propose an automatic feature discovery framework via learning class-specific dictionaries and present a low-complexity method for classification and disease grading in histopathology. Essentially, our Discriminative Feature-oriented Dictionary Learning (DFDL) method learns class-specific dictionaries such that under a sparsity constraint, the learned dictionaries allow representing a new image sample parsimoniously via the dictionary corresponding to the class identity of the sample. At the same time, the dictionary is designed to be poorly capable of representing samples from other classes. Experiments on three challenging real-world image databases: 1) histopathological images of intraductal breast lesions, 2) mammalian kidney, lung and spleen images provided by the Animal Diagnostics Lab (ADL) at Pennsylvania State University, and 3) brain tumor images from The Cancer Genome Atlas (TCGA) database, reveal the merits of our proposal over state-of-the-art alternatives. {Moreover, we demonstrate that DFDL exhibits a more graceful decay in classification accuracy against the number of training images which is highly desirable in practice where generous training is often not available
In multi-label classification, an instance may be associated with a set of labels simultaneously. Recently, the research on multi-label classification has largely shifted its focus to the other end of the spectrum where the number of labels is assumed to be extremely large. The existing works focus on how to design scalable algorithms that offer fast training procedures and have a small memory footprint. However they ignore and even compound another challenge - the label imbalance problem. To address this drawback, we propose a novel Representation-based Multi-label Learning with Sampling (RMLS) approach. To the best of our knowledge, we are the first to tackle the imbalance problem in multi-label classification with many labels. Our experimentations with real-world datasets demonstrate the effectiveness of the proposed approach.
Multi-label classification is a common supervised machine learning problem where each instance is associated with multiple classes. The key challenge in this problem is learning the correlations between the classes. An additional challenge arises when the labels of the training instances are provided by noisy, heterogeneous crowdworkers with unknown qualities. We first assume labels from a perfect source and propose a novel topic model where the present as well as the absent classes generate the latent topics and hence the words. We non-trivially extend our topic model to the scenario where the labels are provided by noisy crowdworkers. Extensive experimentation on real world datasets reveals the superior performance of the proposed model. The proposed model learns the qualities of the annotators as well, even with minimal training data.
We address relation classification in the context of slot filling, the task of finding and evaluating fillers like "Steve Jobs" for the slot X in "X founded Apple". We propose a convolutional neural network which splits the input sentence into three parts according to the relation arguments and compare it to state-of-the-art and traditional approaches of relation classification. Finally, we combine different methods and show that the combination is better than individual approaches. We also analyze the effect of genre differences on performance.
This paper presents a novel latent variable recurrent neural network architecture for jointly modeling sequences of words and (possibly latent) discourse relations between adjacent sentences. A recurrent neural network generates individual words, thus reaping the benefits of discriminatively-trained vector representations. The discourse relations are represented with a latent variable, which can be predicted or marginalized, depending on the task. The resulting model can therefore employ a training objective that includes not only discourse relation classification, but also word prediction. As a result, it outperforms state-of-the-art alternatives for two tasks: implicit discourse relation classification in the Penn Discourse Treebank, and dialog act classification in the Switchboard corpus. Furthermore, by marginalizing over latent discourse relations at test time, we obtain a discourse informed language model, which improves over a strong LSTM baseline.
Learning deeper convolutional neural networks becomes a tendency in recent years. However, many empirical evidences suggest that performance improvement cannot be gained by simply stacking more layers. In this paper, we consider the issue from an information theoretical perspective, and propose a novel method Relay Backpropagation, that encourages the propagation of effective information through the network in training stage. By virtue of the method, we achieved the first place in ILSVRC 2015 Scene Classification Challenge. Extensive experiments on two challenging large scale datasets demonstrate the effectiveness of our method is not restricted to a specific dataset or network architecture. Our models will be available to the research community later.