We propose a novel method to improve deep learning model performance on highly-imbalanced tasks. The proposed method is based on CycleGAN to achieve balanced dataset. We show that data augmentation with GAN helps to improve accuracy of pneumonia binary classification task even if the generative network was trained on the same training dataset.
Simulations are a pedagogical means of enabling a risk-free way for healthcare practitioners to learn, maintain, or enhance their knowledge and skills. Such simulations should provide an optimum amount of cognitive load to the learner and be tailored to their levels of expertise. However, most current simulations are a one-type-fits-all tool used to train different learners regardless of their existing skills, expertise, and ability to handle cognitive load. To address this problem, we propose an end-to-end framework for a trauma simulation that actively classifies a participant's level of cognitive load and expertise for the development of a dynamically adaptive simulation. To facilitate this solution, trauma simulations were developed for the collection of electrocardiogram (ECG) signals of both novice and expert practitioners. A multitask deep neural network was developed to utilize this data and classify high and low cognitive load, as well as expert and novice participants. A leave-one-subject-out (LOSO) validation was used to evaluate the effectiveness of our model, achieving an accuracy of 89.4% and 96.6% for classification of cognitive load and expertise, respectively.
First, we analyze the variance of the Cross Validation (CV)-based estimators used for estimating the performance of classification rules. Second, we propose a novel estimator to estimate this variance using the Influence Function (IF) approach that had been used previously very successfully to estimate the variance of the bootstrap-based estimators. The motivation for this research is that, as the best of our knowledge, the literature lacks a rigorous method for estimating the variance of the CV-based estimators. What is available is a set of ad-hoc procedures that have no mathematical foundation since they ignore the covariance structure among dependent random variables. The conducted experiments show that the IF proposed method has small RMS error with some bias. However, surprisingly, the ad-hoc methods still work better than the IF-based method. Unfortunately, this is due to the lack of enough smoothness if compared to the bootstrap estimator. This opens the research for three points: (1) more comprehensive simulation study to clarify when the IF method win or loose; (2) more mathematical analysis to figure out why the ad-hoc methods work well; and (3) more mathematical treatment to figure out the connection between the appropriate amount of "smoothness" and decreasing the bias of the IF method.
N-grams have been a common tool for information retrieval and machine learning applications for decades. In nearly all previous works, only a few values of $n$ are tested, with $n > 6$ being exceedingly rare. Larger values of $n$ are not tested due to computational burden or the fear of overfitting. In this work, we present a method to find the top-$k$ most frequent $n$-grams that is 60$\times$ faster for small $n$, and can tackle large $n\geq1024$. Despite the unprecedented size of $n$ considered, we show how these features still have predictive ability for malware classification tasks. More important, large $n$-grams provide benefits in producing features that are interpretable by malware analysis, and can be used to create general purpose signatures compatible with industry standard tools like Yara. Furthermore, the counts of common $n$-grams in a file may be added as features to publicly available human-engineered features that rival efficacy of professionally-developed features when used to train gradient-boosted decision tree models on the EMBER dataset.
Object detection and recognition has been an ongoing research topic for a long time in the field of computer vision. Even in robotics, detecting the state of an object by a robot still remains a challenging task. Also, collecting data for each possible state is also not feasible. In this literature, we use a deep convolutional neural network with SVM as a classifier to help with recognizing the state of a cooking object. We also study how a generative adversarial network can be used for synthetic data augmentation and improving the classification accuracy. The main motivation behind this work is to estimate how well a robot could recognize the current state of an object
We revisit the problem of differentially private release of classification queries. In this problem, the goal is to design an algorithm that can accurately answer a sequence of classification queries based on a private training set while ensuring differential privacy. We formally study this problem in the agnostic PAC model and derive a new upper bound on the private sample complexity. Our results improve over those obtained in a recent work [BTT18] for the agnostic PAC setting. In particular, we give an improved construction that yields a tighter upper bound on the sample complexity. Moreover, unlike [BTT18], our accuracy guarantee does not involve any blow-up in the approximation error associated with the given hypothesis class. Given any hypothesis class with VC-dimension $d$, we show that our construction can privately answer up to $m$ classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{m}\,\alpha^{3/2}\right)$. Using recent results on private learning with auxiliary public data, we extend our construction to show that one can privately answer any number of classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{d}\,\alpha\right)$. When $\alpha=O\left(\frac{1}{\sqrt{d}}\right)$, our private sample complexity bound is essentially optimal.
Recent years have witnessed increasing attention in cartoon media, powered by the strong demands of industrial applications. As the first step to understand this media, cartoon face recognition is a crucial but less-explored task with few datasets proposed. In this work, we first present a new challenging benchmark dataset, consisting of 389,678 images of 5,013 cartoon characters annotated with identity, bounding box, pose, and other auxiliary attributes. The dataset, named iCartoonFace, is currently the largest-scale, high-quality, richannotated, and spanning multiple occurrences in the field of image recognition, including near-duplications, occlusions, and appearance changes. In addition, we provide two types of annotations for cartoon media, i.e., face recognition, and face detection, with the help of a semi-automatic labeling algorithm. To further investigate this challenging dataset, we propose a multi-task domain adaptation approach that jointly utilizes the human and cartoon domain knowledge with three discriminative regularizations. We hence perform a benchmark analysis of the proposed dataset and verify the superiority of the proposed approach in the cartoon face recognition task. We believe this public availability will attract more research attention in broad practical application scenarios.
Video Recognition has drawn great research interest and great progress has been made. A suitable frame sampling strategy can improve the accuracy and efficiency of recognition. However, mainstream solutions generally adopt hand-crafted frame sampling strategies for recognition. It could degrade the performance, especially in untrimmed videos, due to the variation of frame-level saliency. To this end, we concentrate on improving untrimmed video classification via developing a learning-based frame sampling strategy. We intuitively formulate the frame sampling procedure as multiple parallel Markov decision processes, each of which aims at picking out a frame/clip by gradually adjusting an initial sampling. Then we propose to solve the problems with multi-agent reinforcement learning (MARL). Our MARL framework is composed of a novel RNN-based context-aware observation network which jointly models context information among nearby agents and historical states of a specific agent, a policy network which generates the probability distribution over a predefined action space at each step and a classification network for reward calculation as well as final recognition. Extensive experimental results show that our MARL-based scheme remarkably outperforms hand-crafted strategies with various 2D and 3D baseline methods. Our single RGB model achieves a comparable performance of ActivityNet v1.3 champion submission with multi-modal multi-model fusion and new state-of-the-art results on YouTube Birds and YouTube Cars.
Multiple classifier system (MCS) has become a successful alternative for improving classification performance. However, studies have shown inconsistent results for different MCSs, and it is often difficult to predict which MCS algorithm works the best on a particular problem. We believe that the two crucial steps of MCS - base classifier generation and multiple classifier combination, need to be designed coordinately to produce robust results. In this work, we show that for different testing instances, better classifiers may be trained from different subdomains of training instances including, for example, neighboring instances of the testing instance, or even instances far away from the testing instance. To utilize this intuition, we propose Individualized Classifier Ensemble (ICE). ICE groups training data into overlapping clusters, builds a classifier for each cluster, and then associates each training instance to the top-performing models while taking into account model types and frequency. In testing, ICE finds the k most similar training instances for a testing instance, then predicts class label of the testing instance by averaging the prediction from models associated with these training instances. Evaluation results on 49 benchmarks show that ICE has a stable improvement on a significant proportion of datasets over existing MCS methods. ICE provides a novel choice of utilizing internal patterns among instances to improve classification, and can be easily combined with various classification models and applied to many application domains.
The development of deep convolutional neural network architecture is critical to the improvement of image classification task performance. A lot of studies of image classification based on deep convolutional neural network focus on the network structure to improve the image classification performance. Contrary to these studies, we focus on the loss function. Cross-entropy Loss (CEL) is widely used for training a multi-class classification deep convolutional neural network. While CEL has been successfully implemented in image classification tasks, it only focuses on the posterior probability of correct class when the labels of training images are one-hot. It cannot be discriminated against the classes not belong to correct class (wrong classes) directly. In order to solve the problem of CEL, we propose Competing Ratio Loss (CRL), which calculates the posterior probability ratio between the correct class and competing wrong classes to better discriminate the correct class from competing wrong classes, increasing the difference between the negative log likelihood of the correct class and the negative log likelihood of competing wrong classes, widening the difference between the probability of the correct class and the probabilities of wrong classes. To demonstrate the effectiveness of our loss function, we perform some sets of experiments on different types of image classification datasets, including CIFAR, SVHN, CUB200- 2011, Adience and ImageNet datasets. The experimental results show the effectiveness and robustness of our loss function on different deep convolutional neural network architectures and different image classification tasks, such as fine-grained image classification, hard face age estimation and large-scale image classification.
General fuzzy min-max (GFMM) neural network is a generalization of fuzzy neural networks formed by hyperbox fuzzy sets for classification and clustering problems. Two principle algorithms are deployed to train this type of neural network, i.e., incremental learning and agglomerative learning. This paper presents a comprehensive empirical study of performance influencing factors, advantages, and drawbacks of the general fuzzy min-max neural network on pattern classification problems. The subjects of this study include (1) the impact of maximum hyperbox size, (2) the influence of the similarity threshold and measures on the agglomerative learning algorithm, (3) the effect of data presentation order, (4) comparative performance evaluation of the GFMM with other types of fuzzy min-max neural networks and prevalent machine learning algorithms. The experimental results on benchmark datasets widely used in machine learning showed overall strong and weak points of the GFMM classifier. These outcomes also informed potential research directions for this class of machine learning algorithms in the future.
Research into automated systems for detecting and classifying marine mammals in acoustic recordings is expanding internationally due to the necessity to analyze large collections of data for conservation purposes. In this work, we present a Convolutional Neural Network that is capable of classifying the vocalizations of three species of whales, non-biological sources of noise, and a fifth class pertaining to ambient noise. In this way, the classifier is capable of detecting the presence and absence of whale vocalizations in an acoustic recording. Through transfer learning, we show that the classifier is capable of learning high-level representations and can generalize to additional species. We also propose a novel representation of acoustic signals that builds upon the commonly used spectrogram representation by way of interpolating and stacking multiple spectrograms produced using different Short-time Fourier Transform (STFT) parameters. The proposed representation is particularly effective for the task of marine mammal species classification where the acoustic events we are attempting to classify are sensitive to the parameters of the STFT.
The information theoretic quantity known as mutual information finds wide use in classification and community detection analyses to compare two classifications of the same set of objects into groups. In the context of classification algorithms, for instance, it is often used to compare discovered classes to known ground truth and hence to quantify algorithm performance. Here we argue that the standard mutual information, as commonly defined, omits a crucial term which can become large under real-world conditions, producing results that can be substantially in error. We demonstrate how to correct this error and define a mutual information that works in all cases. We discuss practical implementation of the new measure and give some example applications.
Action detection and recognition tasks have been the target of much focus in the computer vision community due to their many applications, namely, security, robotics and recommendation systems. Recently, datasets like AVA, provide multi-person, multi-label, spatiotemporal action detection and recognition challenges. Being unable to discern which portions of the input to use for classification is a limitation of two-stream CNN approaches, once the vision task involves several people with several labels. We address this limitation and improve the state-of-the-art performance of two-stream CNNs. In this paper we present four contributions: our fovea attention filtering that highlights targets for classification without discarding background; a generalized binary loss function designed for the AVA dataset; miniAVA, a partition of AVA that maintains temporal continuity and class distribution with only one tenth of the dataset size; and ablation studies on alternative attention filters. Our method, using fovea attention filtering and our generalized binary loss, achieves a relative video mAP improvement of 20% over the two-stream baseline in AVA, and is competitive with the state-of-the-art in the UCF101-24. We also show a relative video mAP improvement of 12.6% when using our generalized binary loss over the standard sum-of-sigmoids.
For the task of concurrently detecting and categorizing objects, the medical imaging community commonly adopts methods developed on natural images. Current state-of-the-art object detectors are comprised of two stages: the first stage generates region proposals, the second stage subsequently categorizes them. Unlike in natural images, however, for anatomical structures of interest such as tumors, the appearance in the image (e.g., scale or intensity) links to a malignancy grade that lies on a continuous ordinal scale. While classification models discard this ordinal relation between grades by discretizing the continuous scale to an unordered bag of categories, regression models are trained with distance metrics, which preserve the relation. This advantage becomes all the more important in the setting of label confusions on ambiguous data sets, which is the usual case with medical images. To this end, we propose Reg R-CNN, which replaces the second-stage classification model of a current object detector with a regression model. We show the superiority of our approach on a public data set with 1026 patients and a series of toy experiments. Code will be available at github.com/MIC-DKFZ/RegRCNN.
An explainable machine learning method for point cloud classification, called the PointHop method, is proposed in this work. The PointHop method consists of two stages: 1) local-to-global attribute building through iterative one-hop information exchange, and 2) classification and ensembles. In the attribute building stage, we address the problem of unordered point cloud data using a space partitioning procedure and developing a robust descriptor that characterizes the relationship between a point and its one-hop neighbor in a PointHop unit. When we put multiple PointHop units in cascade, the attributes of a point will grow by taking its relationship with one-hop neighbor points into account iteratively. Furthermore, to control the rapid dimension growth of the attribute vector associated with a point, we use the Saab transform to reduce the attribute dimension in each PointHop unit. In the classification and ensemble stage, we feed the feature vector obtained from multiple PointHop units to a classifier. We explore ensemble methods to improve the classification performance furthermore. It is shown by experimental results that the PointHop method offers classification performance that is comparable with state-of-the-art methods while demanding much lower training complexity.
Deep Convolutional Neural Networks (CNNs) have been widely used in image classification tasks, but the process of designing CNN architectures is very complex, so Neural Architecture Search (NAS), automatically searching for optimal CNN architectures, has attracted more and more research interests. However, the computational cost of NAS is often too high to apply NAS on real-life applications. In this paper, an efficient particle swarm optimisation method named EPSOCNN is proposed to evolve CNN architectures inspired by the idea of transfer learning. EPSOCNN successfully reduces the computation cost by minimising the search space to a single block and utilising a small subset of the training set to evaluate CNNs during evolutionary process. Meanwhile, EPSOCNN also keeps very competitive classification accuracy by stacking the evolved block multiple times to fit the whole dataset. The proposed EPSOCNN algorithm is evaluated on CIFAR-10 dataset and compared with 13 peer competitors comprised of deep CNNs crafted by hand, learned by reinforcement learning methods and evolved by evolutionary computation approaches, which shows very promising results by outperforming all of the peer competitors with regard to the classification accuracy, number of parameters and the computational cost.
Yarbus' claim to decode the observer's task from eye movements has received mixed reactions. In this paper, we have supported the hypothesis that it is possible to decode the task. We conducted an exploratory analysis on the dataset by projecting features and data points into a scatter plot to visualize the nuance properties for each task. Following this analysis, we eliminated highly correlated features before training an SVM and Ada Boosting classifier to predict the tasks from this filtered eye movements data. We achieve an accuracy of 95.4% on this task classification problem and hence, support the hypothesis that task classification is possible from a user's eye movement data.
MobileNet and Binary Neural Networks are two among the most widely used techniques to construct deep learning models for performing a variety of tasks on mobile and embedded platforms.In this paper, we present a simple yet efficient scheme to exploit MobileNet binarization at activation function and model weights. However, training a binary network from scratch with separable depth-wise and point-wise convolutions in case of MobileNet is not trivial and prone to divergence. To tackle this training issue, we propose a novel neural network architecture, namely MoBiNet - Mobile Binary Network in which skip connections are manipulated to prevent information loss and vanishing gradient, thus facilitate the training process. More importantly, while existing binary neural networks often make use of cumbersome backbones such as Alex-Net, ResNet, VGG-16 with float-type pre-trained weights initialization, our MoBiNet focuses on binarizing the already-compressed neural networks like MobileNet without the need of a pre-trained model to start with. Therefore, our proposal results in an effectively small model while keeping the accuracy comparable to existing ones. Experiments on ImageNet dataset show the potential of the MoBiNet as it achieves 54.40% top-1 accuracy and dramatically reduces the computational cost with binary operators.
Geo-tagged Twitter data has been used recently to infer insights on the human aspects of social media. Insights related to demographics, spatial distribution of cultural activities, space-time travel trajectories for humans as well as happiness has been mined from geo-tagged twitter data in recent studies. To date, not much study has been done on the impact of the geolocation features of a Tweet on its sentiment. This observation has inspired us to propose the usage of geo-location features as a method to perform sentiment classification. In this method, the sentiment classification of geo-tagged tweets is performed by concatenating geo-location features and one-hot encoded word vectors as inputs for convolutional neural networks (CNN) and long short-term memory (LSTM) networks. The addition of language-independent features in the form of geo-location features has helped to enrich the tweet representation in order to combat the sparse nature of short tweet message. The results achieved has demonstrated that concatenating geo-location features to one-hot encoded word vectors can achieve higher accuracy as compared to the usage of word vectors alone for the purpose of sentiment classification.