PyTorchKR Papers

최신 논문

총 11,126편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2202.06369

Incremental user embedding modeling for personalized text classification

Individual user profiles and interaction histories play a significant role in providing customized experiences in real-world applications such as chatbots, social media, retail, and education. Adaptive user representation learning by utilizing user personalized information has become increasingly challenging due to ever-growing history data. In this work, we propose an incremental user embedding modeling approach, in which embeddings of user's recent interaction histories are dynamically integrated into the accumulated history vectors via a transformer encoder. This modeling paradigm allows us to create generalized user representations in a consecutive manner and also alleviate the challenges of data management. We demonstrate the effectiveness of this approach by applying it to a personalized multi-class classification task based on the Reddit dataset, and achieve 9% and 30% relative improvement on prediction accuracy over a baseline system for two experiment settings through appropriate comment history encoding and task modeling.

arXiv:2202.06076공식 코드 · Papers with Code

Indication as Prior Knowledge for Multimodal Disease Classification in Chest Radiographs with Transformers

When a clinician refers a patient for an imaging exam, they include the reason (e.g. relevant patient history, suspected disease) in the scan request; this appears as the indication field in the radiology report. The interpretation and reporting of the image are substantially influenced by this request text, steering the radiologist to focus on particular aspects of the image. We use the indication field to drive better image classification, by taking a transformer network which is unimodally pre-trained on text (BERT) and fine-tuning it for multimodal classification of a dual image-text input. We evaluate the method on the MIMIC-CXR dataset, and present ablation studies to investigate the effect of the indication field on the classification performance. The experimental results show our approach achieves 87.8 average micro AUROC, outperforming the state-of-the-art methods for unimodal (84.4) and multimodal (86.0) classification. Our code is available at https://github.com/jacenkow/mmbt.

arXiv:2202.06600

Research on Dual Channel News Headline Classification Based on ERNIE Pre-training Model

The classification of news headlines is an important direction in the field of NLP, and its data has the characteristics of compactness, uniqueness and various forms. Aiming at the problem that the traditional neural network model cannot adequately capture the underlying feature information of the data and cannot jointly extract key global features and deep local features, a dual-channel network model DC-EBAD based on the ERNIE pre-training model is proposed. Use ERNIE to extract the lexical, semantic and contextual feature information at the bottom of the text, generate dynamic word vector representations fused with context, and then use the BiLSTM-AT network channel to secondary extract the global features of the data and use the attention mechanism to give key parts higher The weight of the DPCNN channel is used to overcome the long-distance text dependence problem and obtain deep local features. The local and global feature vectors are spliced, and finally passed to the fully connected layer, and the final classification result is output through Softmax. The experimental results show that the proposed model improves the accuracy, precision and F1-score of news headline classification compared with the traditional neural network model and the single-channel model under the same conditions. It can be seen that it can perform well in the multi-classification application of news headline text under large data volume.

arXiv:2202.06335공식 코드 · Papers with Code

ET-BERT: A Contextualized Datagram Representation with Pre-training Transformers for Encrypted Traffic Classification

Encrypted traffic classification requires discriminative and robust traffic representation captured from content-invisible and imbalanced traffic data for accurate classification, which is challenging but indispensable to achieve network security and network management. The major limitation of existing solutions is that they highly rely on the deep features, which are overly dependent on data size and hard to generalize on unseen data. How to leverage the open-domain unlabeled traffic data to learn representation with strong generalization ability remains a key challenge. In this paper,we propose a new traffic representation model called Encrypted Traffic Bidirectional Encoder Representations from Transformer (ET-BERT), which pre-trains deep contextualized datagram-level representation from large-scale unlabeled data. The pre-trained model can be fine-tuned on a small number of task-specific labeled data and achieves state-of-the-art performance across five encrypted traffic classification tasks, remarkably pushing the F1 of ISCX-Tor to 99.2% (4.4% absolute improvement), ISCX-VPN-Service to 98.9% (5.2% absolute improvement), Cross-Platform (Android) to 92.5% (5.4% absolute improvement), CSTNET-TLS 1.3 to 97.4% (10.0% absolute improvement). Notably, we provide explanation of the empirically powerful pre-training model by analyzing the randomness of ciphers. It gives us insights in understanding the boundary of classification ability over encrypted traffic. The code is available at: https://github.com/linwhitehat/ET-BERT.

arXiv:2202.05397

Neural Architecture Search for Energy Efficient Always-on Audio Models

Mobile and edge computing devices for always-on classification tasks require energy-efficient neural network architectures. In this paper we present several changes to neural architecture searches (NAS) that improve the chance of success in practical situations. Our search simultaneously optimizes for network accuracy, energy efficiency and memory usage. We benchmark the performance of our search on real hardware, but since running thousands of tests with real hardware is difficult we use a random forest model to roughly predict the energy usage of a candidate network. We present a search strategy that uses both Bayesian and regularized evolutionary search with particle swarms, and employs early-stopping to reduce the computational burden. Our search, evaluated on a sound-event classification dataset based upon AudioSet, results in an order of magnitude less energy per inference and a much smaller memory footprint than our baseline MobileNetV1/V2 implementations while slightly improving task accuracy. We also demonstrate how combining a 2D spectrogram with a convolution with many filters causes a computational bottleneck for audio classification and that alternative approaches reduce the computational burden but sacrifice task accuracy.

arXiv:2202.05685

SuperCon: Supervised Contrastive Learning for Imbalanced Skin Lesion Classification

Convolutional neural networks (CNNs) have achieved great success in skin lesion classification. A balanced dataset is required to train a good model. However, due to the appearance of different skin lesions in practice, severe or even deadliest skin lesion types (e.g., melanoma) naturally have quite small amount represented in a dataset. In that, classification performance degradation occurs widely, it is significantly important to have CNNs that work well on class imbalanced skin lesion image dataset. In this paper, we propose SuperCon, a two-stage training strategy to overcome the class imbalance problem on skin lesion classification. It contains two stages: (i) representation training that tries to learn a feature representation that closely aligned among intra-classes and distantly apart from inter-classes, and (ii) classifier fine-tuning that aims to learn a classifier that correctly predict the label based on the learnt representations. In the experimental evaluation, extensive comparisons have been made among our approach and other existing approaches on skin lesion benchmark datasets. The results show that our two-stage training strategy effectively addresses the class imbalance classification problem, and significantly improves existing works in terms of F1-score and AUC score, resulting in state-of-the-art performance.

arXiv:2202.05560

Controlling Multiple Errors Simultaneously with a PAC-Bayes Bound

Current PAC-Bayes generalisation bounds are restricted to scalar metrics of performance, such as the loss or error rate. However, one ideally wants more information-rich certificates that control the entire distribution of possible outcomes, such as the distribution of the test loss in regression, or the probabilities of different mis-classifications. We provide the first PAC-Bayes bound capable of providing such rich information by bounding the Kullback-Leibler divergence between the empirical and true probabilities of a set of $M$ error types, which can either be discretized loss values for regression, or the elements of the confusion matrix (or a partition thereof) for classification. We transform our bound into a differentiable training objective. Our bound is especially useful in cases where the severity of different mis-classifications may change over time; existing PAC-Bayes bounds can only bound a particular pre-decided weighting of the error types. In contrast our bound implicitly controls all uncountably many weightings simultaneously.

arXiv:2202.05339

Closure operators: Complexity and applications to classification and decision-making

We study the complexity of closure operators, with applications to machine learning and decision theory. In machine learning, closure operators emerge naturally in data classification and clustering. In decision theory, they can model equivalence of choice menus, and therefore situations with a preference for flexibility. Our contribution is to formulate a notion of complexity of closure operators, which translate into the complexity of a classifier in ML, or of a utility function in decision theory.

논문 ID: document/9197689

Fuzzy Pooling

Convolutional Neural Networks (CNNs) are artificial learning systems typically based on two operations: convolution, which implements feature extraction through filtering, and pooling, which implements dimensionality reduction. The impact of pooling in the classification performance of the CNNs has been highlighted in several previous works, and a variety of alternative pooling operators have been proposed. However, only a few of them tackle with the uncertainty that is naturally propagated from the input layer to the feature maps of the hidden layers through convolutions. In this paper we present a novel pooling operation based on (type-1) fuzzy sets to cope with the local imprecision of the feature maps, and we investigate its performance in the context of image classification. Fuzzy pooling is performed by fuzzification, aggregation and defuzzification of feature map neighborhoods. It is used for the construction of a fuzzy pooling layer that can be applied as a drop-in replacement of the current, crisp, pooling layers of CNN architectures. Several experiments using publicly available datasets show that the proposed approach can enhance the classification performance of a CNN. A comparative evaluation shows that it outperforms state-of-the-art pooling approaches.

논문 ID: document/9145825

C-CNN: Contourlet Convolutional Neural Networks

Extracting effective features is always a challenging problem for texture classification because of the uncertainty of scales and the clutter of textural patterns. For texture classification, spectral analysis is traditionally employed in the frequency domain. Recent studies have shown the potential of convolutional neural networks (CNNs) when dealing with the texture classification task in the spatial domain. In this article, we try combining both approaches in different domains for more abundant information and proposed a novel network architecture named contourlet CNN (C-CNN). The network aims to learn sparse and effective feature representations for images. First, the contourlet transform is applied to get the spectral features from an image. Second, the spatial-spectral feature fusion strategy is designed to incorporate the spectral features into CNN architecture. Third, the statistical features are integrated into the network by the statistical feature fusion. Finally, the results are obtained by classifying the fusion features. We also investigated the behavior of the parameters in contourlet decomposition. Experiments on the widely used three texture data sets (kth-tips2-b, DTD, and CUReT) and five remote sensing data sets (UCM, WHU-RS, AID, RSSCN7, and NWPU-RESISC45) demonstrate that the proposed approach outperforms several well-known classification methods in terms of classification accuracy with fewer trainable parameters.

arXiv:2202.05656공식 코드 · Papers with Code

Evaluation of post-hoc interpretability methods in time-series classification

Post-hoc interpretability methods are critical tools to explain neural-network results. Several post-hoc methods have emerged in recent years, but when applied to a given task, they produce different results, raising the question of which method is the most suitable to provide correct post-hoc interpretability. To understand the performance of each method, quantitative evaluation of interpretability methods is essential. However, currently available frameworks have several drawbacks which hinders the adoption of post-hoc interpretability methods, especially in high-risk sectors. In this work, we propose a framework with quantitative metrics to assess the performance of existing post-hoc interpretability methods in particular in time series classification. We show that several drawbacks identified in the literature are addressed, namely dependence on human judgement, retraining, and shift in the data distribution when occluding samples. We additionally design a synthetic dataset with known discriminative features and tunable complexity. The proposed methodology and quantitative metrics can be used to understand the reliability of interpretability methods results obtained in practical applications. In turn, they can be embedded within operational workflows in critical fields that require accurate interpretability results for e.g., regulatory policies.

arXiv:2202.05169

Radar-based Materials Classification Using Deep Wavelet Scattering Transform: A Comparison of Centimeter vs. Millimeter Wave Units

Radar-based materials detection received significant attention in recent years for its potential inclusion in consumer and industrial applications like object recognition for grasping and manufacturing quality assurance and control. Several radar publications were developed for material classification under controlled settings with specific materials' properties and shapes. Recent literature has challenged the earlier findings on radars-based materials classification claiming that earlier solutions are not easily scaled to industrial applications due to a variety of real-world issues. Published experiments on the impact of these factors on the robustness of the extracted radar-based traditional features have already demonstrated that the application of deep neural networks can mitigate, to some extent, the impact to produce a viable solution. However, previous studies lacked an investigation of the usefulness of lower frequency radar units, specifically <10GHz, against the higher range units around and above 60GHz. This research considers two radar units with different frequency ranges: Walabot-3D (6.3-8 GHz) cm-wave and IMAGEVK-74 (62-69 GHz) mm-wave imaging units by Vayyar Imaging. A comparison is presented on the applicability of each unit for material classification. This work extends upon previous efforts, by applying deep wavelet scattering transform for the identification of different materials based on the reflected signals. In the wavelet scattering feature extractor, data is propagated through a series of wavelet transforms, nonlinearities, and averaging to produce low-variance representations of the reflected radar signals. This work is unique in comparison of the radar units and algorithms in material classification and includes real-time demonstrations that show strong performance by both units, with increased robustness offered by the cm-wave radar unit.

arXiv:2202.04844공식 코드 · Papers with Code

Multi-relation Message Passing for Multi-label Text Classification

A well-known challenge associated with the multi-label classification problem is modelling dependencies between labels. Most attempts at modelling label dependencies focus on co-occurrences, ignoring the valuable information that can be extracted by detecting label subsets that rarely occur together. For example, consider customer product reviews; a product probably would not simultaneously be tagged by both "recommended" (i.e., reviewer is happy and recommends the product) and "urgent" (i.e., the review suggests immediate action to remedy an unsatisfactory experience). Aside from the consideration of positive and negative dependencies, the direction of a relationship should also be considered. For a multi-label image classification problem, the "ship" and "sea" labels have an obvious dependency, but the presence of the former implies the latter much more strongly than the other way around. These examples motivate the modelling of multiple types of bi-directional relationships between labels. In this paper, we propose a novel method, entitled Multi-relation Message Passing (MrMP), for the multi-label classification problem. Experiments on benchmark multi-label text classification datasets show that the MrMP module yields similar or superior performance compared to state-of-the-art methods. The approach imposes only minor additional computational and memory overheads.

arXiv:2202.04769

Spectral Propagation Graph Network for Few-shot Time Series Classification

Few-shot Time Series Classification (few-shot TSC) is a challenging problem in time series analysis. It is more difficult to classify when time series of the same class are not completely consistent in spectral domain or time series of different classes are partly consistent in spectral domain. To address this problem, we propose a novel method named Spectral Propagation Graph Network (SPGN) to explicitly model and propagate the spectrum-wise relations between different time series with graph network. To the best of our knowledge, SPGN is the first to utilize spectral comparisons in different intervals and involve spectral propagation across all time series with graph networks for few-shot TSC. SPGN first uses bandpass filter to expand time series in spectral domain for calculating spectrum-wise relations between time series. Equipped with graph networks, SPGN then integrates spectral relations with label information to make spectral propagation. The further study conveys the bi-directional effect between spectral relations acquisition and spectral propagation. We conduct extensive experiments on few-shot TSC benchmarks. SPGN outperforms state-of-the-art results by a large margin in $4\% \sim 13\%$. Moreover, SPGN surpasses them by around $12\%$ and $9\%$ under cross-domain and cross-way settings respectively.

arXiv:2202.05246

Monotone Learning

The amount of training-data is one of the key factors which determines the generalization capacity of learning algorithms. Intuitively, one expects the error rate to decrease as the amount of training-data increases. Perhaps surprisingly, natural attempts to formalize this intuition give rise to interesting and challenging mathematical questions. For example, in their classical book on pattern recognition, Devroye, Gyorfi, and Lugosi (1996) ask whether there exists a {monotone} Bayes-consistent algorithm. This question remained open for over 25 years, until recently Pestov (2021) resolved it for binary classification, using an intricate construction of a monotone Bayes-consistent algorithm. We derive a general result in multiclass classification, showing that every learning algorithm A can be transformed to a monotone one with similar performance. Further, the transformation is efficient and only uses a black-box oracle access to A. This demonstrates that one can provably avoid non-monotonic behaviour without compromising performance, thus answering questions asked by Devroye et al (1996), Viering, Mey, and Loog (2019), Viering and Loog (2021), and by Mhammedi (2021). Our transformation readily implies monotone learners in a variety of contexts: for example it extends Pestov's result to classification tasks with an arbitrary number of labels. This is in contrast with Pestov's work which is tailored to binary classification. In addition, we provide uniform bounds on the error of the monotone algorithm. This makes our transformation applicable in distribution-free settings. For example, in PAC learning it implies that every learnable class admits a monotone PAC learner. This resolves questions by Viering, Mey, and Loog (2019); Viering and Loog (2021); Mhammedi (2021).

arXiv:2202.05170

Efficacy of Transformer Networks for Classification of Raw EEG Data

With the unprecedented success of transformer networks in natural language processing (NLP), recently, they have been successfully adapted to areas like computer vision, generative adversarial networks (GAN), and reinforcement learning. Classifying electroencephalogram (EEG) data has been challenging and researchers have been overly dependent on pre-processing and hand-crafted feature extraction. Despite having achieved automated feature extraction in several other domains, deep learning has not yet been accomplished for EEG. In this paper, the efficacy of the transformer network for the classification of raw EEG data (cleaned and pre-processed) is explored. The performance of transformer networks was evaluated on a local (age and gender data) and a public dataset (STEW). First, a classifier using a transformer network is built to classify the age and gender of a person with raw resting-state EEG data. Second, the classifier is tuned for mental workload classification with open access raw multi-tasking mental workload EEG data (STEW). The network achieves an accuracy comparable to state-of-the-art accuracy on both the local (Age and Gender dataset; 94.53% (gender) and 87.79% (age)) and the public (STEW dataset; 95.28% (two workload levels) and 88.72% (three workload levels)) dataset. The accuracy values have been achieved using raw EEG data without feature extraction. Results indicate that the transformer-based deep learning models can successfully abate the need for heavy feature-extraction of EEG data for successful classification.

arXiv:2202.04812

Weakly-Supervised Semantic Segmentation with Visual Words Learning and Hybrid Pooling

Weakly-Supervised Semantic Segmentation (WSSS) methods with image-level labels generally train a classification network to generate the Class Activation Maps (CAMs) as the initial coarse segmentation labels. However, current WSSS methods still perform far from satisfactorily because their adopted CAMs 1) typically focus on partial discriminative object regions and 2) usually contain useless background regions. These two problems are attributed to the sole image-level supervision and aggregation of global information when training the classification networks. In this work, we propose the visual words learning module and hybrid pooling approach, and incorporate them in the classification network to mitigate the above problems. In the visual words learning module, we counter the first problem by enforcing the classification network to learn fine-grained visual word labels so that more object extents could be discovered. Specifically, the visual words are learned with a codebook, which could be updated via two proposed strategies, i.e. learning-based strategy and memory-bank strategy. The second drawback of CAMs is alleviated with the proposed hybrid pooling, which incorporates the global average and local discriminative information to simultaneously ensure object completeness and reduce background regions. We evaluated our methods on PASCAL VOC 2012 and MS COCO 2014 datasets. Without any extra saliency prior, our method achieved 70.6% and 70.7% mIoU on the $val$ and $test$ set of PASCAL VOC dataset, respectively, and 36.2% mIoU on the $val$ set of MS COCO dataset, which significantly surpassed the performance of state-of-the-art WSSS methods.

arXiv:2202.04303

TinyM$^2$Net: A Flexible System Algorithm Co-designed Multimodal Learning Framework for Tiny Devices

With the emergence of Artificial Intelligence (AI), new attention has been given to implement AI algorithms on resource constrained tiny devices to expand the application domain of IoT. Multimodal Learning has recently become very popular with the classification task due to its impressive performance for both image and audio event classification. This paper presents TinyM$^2$Net -- a flexible system algorithm co-designed multimodal learning framework for resource constrained tiny devices. The framework was designed to be evaluated on two different case-studies: COVID-19 detection from multimodal audio recordings and battle field object detection from multimodal images and audios. In order to compress the model to implement on tiny devices, substantial network architecture optimization and mixed precision quantization were performed (mixed 8-bit and 4-bit). TinyM$^2$Net shows that even a tiny multimodal learning model can improve the classification performance than that of any unimodal frameworks. The most compressed TinyM$^2$Net achieves 88.4% COVID-19 detection accuracy (14.5% improvement from unimodal base model) and 96.8% battle field object detection accuracy (3.9% improvement from unimodal base model). Finally, we test our TinyM$^2$Net models on a Raspberry Pi 4 to see how they perform when deployed to a resource constrained tiny device.

arXiv:2202.04357공식 코드 · Papers with Code

Generalized Strategic Classification and the Case of Aligned Incentives

Strategic classification studies learning in settings where self-interested users can strategically modify their features to obtain favorable predictive outcomes. A key working assumption, however, is that "favorable" always means "positive"; this may be appropriate in some applications (e.g., loan approval), but reduces to a fairly narrow view of what user interests can be. In this work we argue for a broader perspective on what accounts for strategic user behavior, and propose and study a flexible model of generalized strategic classification. Our generalized model subsumes most current models but includes other novel settings; among these, we identify and target one intriguing sub-class of problems in which the interests of users and the system are aligned. This setting reveals a surprising fact: that standard max-margin losses are ill-suited for strategic inputs. Returning to our fully generalized model, we propose a novel max-margin framework for strategic learning that is practical and effective, and which we analyze theoretically. We conclude with a set of experiments that empirically demonstrate the utility of our approach.

arXiv:2202.03898

Verification-Aided Deep Ensemble Selection

Deep neural networks (DNNs) have become the technology of choice for realizing a variety of complex tasks. However, as highlighted by many recent studies, even an imperceptible perturbation to a correctly classified input can lead to misclassification by a DNN. This renders DNNs vulnerable to strategic input manipulations by attackers, and also oversensitive to environmental noise. To mitigate this phenomenon, practitioners apply joint classification by an *ensemble* of DNNs. By aggregating the classification outputs of different individual DNNs for the same input, ensemble-based classification reduces the risk of misclassifications due to the specific realization of the stochastic training process of any single DNN. However, the effectiveness of a DNN ensemble is highly dependent on its members *not simultaneously erring* on many different inputs. In this case study, we harness recent advances in DNN verification to devise a methodology for identifying ensemble compositions that are less prone to simultaneous errors, even when the input is adversarially perturbed -- resulting in more robustly-accurate ensemble-based classification. Our proposed framework uses a DNN verifier as a backend, and includes heuristics that help reduce the high complexity of directly verifying ensembles. More broadly, our work puts forth a novel universal objective for formal verification that can potentially improve the robustness of real-world, deep-learning-based systems across a variety of application domains.