Generative Adversarial Networks (GANs) have shown promise in augmenting datasets and boosting convolutional neural networks' (CNN) performance on image classification tasks. But they introduce more hyperparameters to tune as well as the need for additional time and computational power to train supplementary to the CNN. In this work, we examine the potential for Auxiliary-Classifier GANs (AC-GANs) as a 'one-stop-shop' architecture for image classification, particularly in low data regimes. Additionally, we explore modifications to the typical AC-GAN framework, changing the generator's latent space sampling scheme and employing a Wasserstein loss with gradient penalty to stabilize the simultaneous training of image synthesis and classification. Through experiments on images of varying resolutions and complexity, we demonstrate that AC-GANs show promise in image classification, achieving competitive performance with standard CNNs. These methods can be employed as an 'all-in-one' framework with particular utility in the absence of large amounts of training data.
In astronomical surveys, such as the Zwicky Transient Facility, supernovae (SNe) are relatively uncommon objects compared to other classes of variable events. Along with this scarcity, the processing of multi-band light-curves is a challenging task due to the highly irregular cadence, long time gaps, missing-values, few observations, etc. These issues are particularly detrimental to the analysis of transient events: SN-like light-curves. We offer three main contributions: 1) Based on temporal modulation and attention mechanisms, we propose a Deep attention model (TimeModAttn) to classify multi-band light-curves of different SN types, avoiding photometric or hand-crafted feature computations, missing-value assumptions, and explicit imputation/interpolation methods. 2) We propose a model for the synthetic generation of SN multi-band light-curves based on the Supernova Parametric Model, allowing us to increase the number of samples and the diversity of cadence. Thus, the TimeModAttn model is first pre-trained using synthetic light-curves. Then, a fine-tuning process is performed. The TimeModAttn model outperformed other Deep Learning models, based on Recurrent Neural Networks, in two scenarios: late-classification and early-classification. Also, the TimeModAttn model outperformed a Balanced Random Forest (BRF) classifier (trained with real data), increasing the balanced-$F_1$score from $\approx.525$ to $\approx.596$. When training the BRF with synthetic data, this model achieved similar performance to the TimeModAttn model proposed while still maintaining extra advantages. 3) We conducted interpretability experiments. High attention scores were obtained for observations earlier than and close to the SN brightness peaks. This also correlated with an early highly variability of the learned temporal modulation.
Contrastive learning has achieved remarkable success in representation learning via self-supervision in unsupervised settings. However, effectively adapting contrastive learning to supervised learning tasks remains as a challenge in practice. In this work, we introduce a dual contrastive learning (DualCL) framework that simultaneously learns the features of input samples and the parameters of classifiers in the same space. Specifically, DualCL regards the parameters of the classifiers as augmented samples associating to different labels and then exploits the contrastive learning between the input samples and the augmented samples. Empirical studies on five benchmark text classification datasets and their low-resource version demonstrate the improvement in classification accuracy and confirm the capability of learning discriminative representations of DualCL.
Recently, deep learning has achieved remarkable successes in medical image analysis. Although deep neural networks generate clinically important predictions, they have inherent uncertainty. Such uncertainty is a major barrier to report these predictions with confidence. In this paper, we propose a novel yet simple Bayesian inference approach called SoftDropConnect (SDC) to quantify the network uncertainty in medical imaging tasks with gliomas segmentation and metastases classification as initial examples. Our key idea is that during training and testing SDC modulates network parameters continuously so as to allow affected information processing channels still in operation, instead of disabling them as Dropout or DropConnet does. When compared with three popular Bayesian inference methods including Bayes By Backprop, Dropout, and DropConnect, our SDC method (SDC-W after optimization) outperforms the three competing methods with a substantial margin. Quantitatively, our proposed method generates substantial improvements in prediction accuracy (by 3.4%, 2.5%, and 6.7% respectively for whole tumor segmentation in terms of dice score; and by 11.7%, 3.9%, and 8.7% respectively for brain metastases classification) and greatly reduced epistemic and aleatoric uncertainties. Our approach promises to deliver better diagnostic performance and make medical AI imaging more explainable and trustworthy.
We study the problem of few-shot graph classification across domains with nonequivalent feature spaces by introducing three new cross-domain benchmarks constructed from publicly available datasets. We also propose an attention-based graph encoder that uses three congruent views of graphs, one contextual and two topological views, to learn representations of task-specific information for fast adaptation, and task-agnostic information for knowledge transfer. We run exhaustive experiments to evaluate the performance of contrastive and meta-learning strategies. We show that when coupled with metric-based meta-learning frameworks, the proposed encoder achieves the best average meta-test classification accuracy across all benchmarks. The source code and data will be released here: https://github.com/kavehhassani/metagrl
We propose the task of automatically identifying papers used as baselines in a scientific article. We frame the problem as a binary classification task where all the references in a paper are to be classified as either baselines or non-baselines. This is a challenging problem due to the numerous ways in which a baseline reference can appear in a paper. We develop a dataset of $2,075$ papers from ACL anthology corpus with all their references manually annotated as one of the two classes. We develop a multi-module attention-based neural classifier for the baseline classification task that outperforms four state-of-the-art citation role classification methods when applied to the baseline classification task. We also present an analysis of the errors made by the proposed classifier, eliciting the challenges that make baseline identification a challenging problem.
Network and system security are incredibly critical issues now. Due to the rapid proliferation of malware, traditional analysis methods struggle with enormous samples. In this paper, we propose four easy-to-extract and small-scale features, including sizes and permissions of Windows PE sections, content complexity, and import libraries, to classify malware families, and use automatic machine learning to search for the best model and hyper-parameters for each feature and their combinations. Compared with detailed behavior-related features like API sequences, proposed features provide macroscopic information about malware. The analysis is based on static disassembly scripts and hexadecimal machine code. Unlike dynamic behavior analysis, static analysis is resource-efficient and offers complete code coverage, but is vulnerable to code obfuscation and encryption. The results demonstrate that features which work well in dynamic analysis are not necessarily effective when applied to static analysis. For instance, API 4-grams only achieve 57.96% accuracy and involve a relatively high dimensional feature set (5000 dimensions). In contrast, the novel proposed features together with a classical machine learning algorithm (Random Forest) presents very good accuracy at 99.40% and the feature vector is of much smaller dimension (40 dimensions). We demonstrate the effectiveness of this approach through integration in IDA Pro, which also facilitates the collection of new training samples and subsequent model retraining.
Multimodal data arise in various applications where information about the same phenomenon is acquired from multiple sensors and across different imaging modalities. Learning from multimodal data is of great interest in machine learning and statistics research as this offers the possibility of capturing complementary information among modalities. Multimodal modeling helps to explain the interdependence between heterogeneous data sources, discovers new insights that may not be available from a single modality, and improves decision-making. Recently, coupled matrix-tensor factorization has been introduced for multimodal data fusion to jointly estimate latent factors and identify complex interdependence among the latent factors. However, most of the prior work on coupled matrix-tensor factors focuses on unsupervised learning and there is little work on supervised learning using the jointly estimated latent factors. This paper considers the multimodal tensor data classification problem. A Coupled Support Tensor Machine (C-STM) built upon the latent factors jointly estimated from the Advanced Coupled Matrix Tensor Factorization (ACMTF) is proposed. C-STM combines individual and shared latent factors with multiple kernels and estimates a maximal-margin classifier for coupled matrix tensor data. The classification risk of C-STM is shown to converge to the optimal Bayes risk, making it a statistically consistent rule. C-STM is validated through simulation studies as well as a simultaneous EEG-fMRI analysis. The empirical evidence shows that C-STM can utilize information from multiple sources and provide a better classification performance than traditional single-mode classifiers.
Spiking neural networks (SNNs) in neuromorphic systems are more energy efficient compared to deep learning-based methods, but there is no clear competitive learning algorithm for training such SNNs. Eligibility propagation (e-prop) offers an efficient and biologically plausible way to train competitive recurrent SNNs in low-power neuromorphic hardware. In this report, previous performance of e-prop on a speech classification task is reproduced, and the effects of including STDP-like behavior are analyzed. Including STDP to the ALIF neuron model improves the classification performance, but this is not the case for the Izhikevich e-prop neuron. Finally, it was found that e-prop implemented in a single-layer recurrent SNN consistently outperforms a multi-layer variant.
Behavioral Indicators of Compromise are associated with various automated methods used to extract the sample behavior by observing the system function calls performed in a virtual execution environment. Thus, every sample is described by a set of BICs triggered by the sample behavior in the sandbox environment. Here we discuss a Machine Learning approach to the classification of the sandbox samples as MALICIOUS or BENIGN, based on the list of triggered BICs. Besides the more traditional methods like Logistic Regression and Naive Bayes Classification we also discuss a different approach inspired by the statistical Monte Carlo methods. The numerical results are illustrated using ThreatGRID and ReversingLabs data.
Modern single-cell flow and mass cytometry technologies measure the expression of several proteins of the individual cells within a blood or tissue sample. Each profiled biological sample is thus represented by a set of hundreds of thousands of multidimensional cell feature vectors, which incurs a high computational cost to predict each biological sample's associated phenotype with machine learning models. Such a large set cardinality also limits the interpretability of machine learning models due to the difficulty in tracking how each individual cell influences the ultimate prediction. We propose using Kernel Mean Embedding to encode the cellular landscape of each profiled biological sample. Although our foremost goal is to make a more transparent model, we find that our method achieves comparable or better accuracies than the state-of-the-art gating-free methods through a simple linear classifier. As a result, our model contains few parameters but still performs similarly to deep learning models with millions of parameters. In contrast with deep learning approaches, the linearity and sub-selection step of our model makes it easy to interpret classification results. Analysis further shows that our method admits rich biological interpretability for linking cellular heterogeneity to clinical phenotype.
Fake news detection research is still in the early stage as this is a relatively new phenomenon in the interest raised by society. Machine learning helps to solve complex problems and to build AI systems nowadays and especially in those cases where we have tacit knowledge or the knowledge that is not known. We used machine learning algorithms and for identification of fake news; we applied three classifiers; Passive Aggressive, Na\"ive Bayes, and Support Vector Machine. Simple classification is not completely correct in fake news detection because classification methods are not specialized for fake news. With the integration of machine learning and text-based processing, we can detect fake news and build classifiers that can classify the news data. Text classification mainly focuses on extracting various features of text and after that incorporating those features into classification. The big challenge in this area is the lack of an efficient way to differentiate between fake and non-fake due to the unavailability of corpora. We applied three different machine learning classifiers on two publicly available datasets. Experimental analysis based on the existing dataset indicates a very encouraging and improved performance.
Positive-unlabeled learning (PU learning) is known as a special case of semi-supervised binary classification where only a fraction of positive examples are labeled. The challenge is then to find the correct classifier despite this lack of information. Recently, new methodologies have been introduced to address the case where the probability of being labeled may depend on the covariates. In this paper, we are interested in establishing risk bounds for PU learning under this general assumption. In addition, we quantify the impact of label noise on PU learning compared to standard classification setting. Finally, we provide a lower bound on minimax risk proving that the upper bound is almost optimal.
Text classification algorithms investigate the intricate relationships between words or phrases and attempt to deduce the document's interpretation. In the last few years, these algorithms have progressed tremendously. Transformer architecture and sentence encoders have proven to give superior results on natural language processing tasks. But a major limitation of these architectures is their applicability for text no longer than a few hundred words. In this paper, we explore hierarchical transfer learning approaches for long document classification. We employ pre-trained Universal Sentence Encoder (USE) and Bidirectional Encoder Representations from Transformers (BERT) in a hierarchical setup to capture better representations efficiently. Our proposed models are conceptually simple where we divide the input data into chunks and then pass this through base models of BERT and USE. Then output representation for each chunk is then propagated through a shallow neural network comprising of LSTMs or CNNs for classifying the text data. These extensions are evaluated on 6 benchmark datasets. We show that USE + CNN/LSTM performs better than its stand-alone baseline. Whereas the BERT + CNN/LSTM performs on par with its stand-alone counterpart. However, the hierarchical BERT models are still desirable as it avoids the quadratic complexity of the attention mechanism in BERT. Along with the hierarchical approaches, this work also provides a comparison of different deep learning algorithms like USE, BERT, HAN, Longformer, and BigBird for long document classification. The Longformer approach consistently performs well on most of the datasets.
Supervised classification techniques use training samples to learn a classification rule with small expected 0-1 loss (error probability). Conventional methods enable tractable learning and provide out-of-sample generalization by using surrogate losses instead of the 0-1 loss and considering specific families of rules (hypothesis classes). This paper presents minimax risk classifiers (MRCs) that minize the worst-case 0-1 loss with respect to uncertainty sets of distributions that can include the underlying distribution, with a tunable confidence. We show that MRCs can provide tight performance guarantees at learning and are strongly universally consistent using feature mappings given by characteristic kernels. The paper also proposes efficient optimization techniques for MRC learning and shows that the methods presented can provide accurate classification together with tight performance guarantees in practice.
A widely used paradigm to improve the generalization performance of high-capacity neural models is through the addition of auxiliary unsupervised tasks during supervised training. Tasks such as similarity matching and input reconstruction have been shown to provide a beneficial regularizing effect by guiding representation learning. Real data often has complex underlying structures and may be composed of heterogeneous subpopulations that are not learned well with current approaches. In this work, we design ExpertNet, which uses novel training strategies to learn clustered latent representations and leverage them by effectively combining cluster-specific classifiers. We theoretically analyze the effect of clustering on its generalization gap, and empirically show that clustered latent representations from ExpertNet lead to disentangling the intrinsic structure and improvement in classification performance. ExpertNet also meets an important real-world need where classifiers need to be tailored for distinct subpopulations, such as in clinical risk models. We demonstrate the superiority of ExpertNet over state-of-the-art methods on 6 large clinical datasets, where our approach leads to valuable insights on group-specific risks.
In the paper a method for automatic classification of signals received by EKB and MAGW ISTP SB RAS coherent scatter radars (8-20MHz operating frequency) during 2021 is described. The method is suitable for automatic physical interpretation of the resulting classification of the experimental data in realtime. We called this algorithm Wrapped Classifier with Dummy Teacher. The method is trained on unlabeled dataset and is based on training optimal physics-based classification using clusterization results. The approach is close to optimal embedding search, where the embedding is interpreted as a vector of probabilities for soft classification. The approach allows to find optimal classification algorithm, based on physically interpretable parameters of the received data, both obtained during physics-based numerical simulation and measured experimentally. Dummy Teacher clusterer used for labeling unlabeled dataset is gaussian mixture clustering algorithm. For algorithm functioning we extended the parameters obtained by the radar with additional parameters, calculated during simulation of radiowave propagation using ray-tracing and IRI-2012 and IGRF models for ionosphere and Earth's magnetic field correspondingly. For clustering by Dummy Teacher we use the whole dataset of available parameters (measured and simulated ones). For classification by Wrapped Classifier we use only well physically interpreted parameters. As a result we trained the classification network and found 11 well-interpretable classes from physical point of view in the available data. Five other found classes are not interpretable from physical point of view, demonstrating the importance of taking into account radiowave propagation for correct classification.
With this work we are explaining the "You Only Look Once" (YOLO) single-stage object detection approach as a parallel classification of 10647 fixed region proposals. We support this view by showing that each of YOLOs output pixel is attentive to a specific sub-region of previous layers, comparable to a local region proposal. This understanding reduces the conceptual gap between YOLO-like single-stage object detection models, RCNN-like two-stage region proposal based models, and ResNet-like image classification models. In addition, we created interactive exploration tools for a better visual understanding of the YOLO information processing streams: https://limchr.github.io/yolo_visualization
Quick and accurate diagnosis is of paramount importance to mitigate the effects of COVID-19 infection, particularly for severe cases. Enormous effort has been put towards developing deep learning methods to classify and detect COVID-19 infections from chest radiography images. However, recently some questions have been raised surrounding the clinical viability and effectiveness of such methods. In this work, we investigate the impact of multi-task learning (classification and segmentation) on the ability of CNNs to differentiate between various appearances of COVID-19 infections in the lung. We also employ self-supervised pre-training approaches, namely MoCo and inpainting-CXR, to eliminate the dependence on expensive ground truth annotations for COVID-19 classification. Finally, we conduct a critical evaluation of the models to assess their deploy-readiness and provide insights into the difficulties of fine-grained COVID-19 multi-class classification from chest X-rays.
Representation knowledge distillation aims at transferring rich information from one model to another. Common approaches for representation distillation mainly focus on the direct minimization of distance metrics between the models' embedding vectors. Such direct methods may be limited in transferring high-order dependencies embedded in the representation vectors, or in handling the capacity gap between the teacher and student models. Moreover, in standard knowledge distillation, the teacher is trained without awareness of the student's characteristics and capacity. In this paper, we explore two mechanisms for enhancing representation distillation using classifier sharing between the teacher and student. We first investigate a simple scheme where the teacher's classifier is connected to the student backbone, acting as an additional classification head. Then, we propose a student-aware mechanism that asks to tailor the teacher model to a student with limited capacity by training the teacher with a temporary student's head. We analyze and compare these two mechanisms and show their effectiveness on various datasets and tasks, including image classification, fine-grained classification, and face verification. In particular, we achieve state-of-the-art results for face verification on the IJB-C dataset for a MobileFaceNet model: TAR@(FAR=1e-5)=93.7\%. Code is available at https://github.com/Alibaba-MIIL/HeadSharingKD.