PyTorchKR Papers

최신 논문

총 12,350편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1812.01278

Singing Voice Separation Using a Deep Convolutional Neural Network Trained by Ideal Binary Mask and Cross Entropy

Separating a singing voice from its music accompaniment remains an important challenge in the field of music information retrieval. We present a unique neural network approach inspired by a technique that has revolutionized the field of vision: pixel-wise image classification, which we combine with cross entropy loss and pretraining of the CNN as an autoencoder on singing voice spectrograms. The pixel-wise classification technique directly estimates the sound source label for each time-frequency (T-F) bin in our spectrogram image, thus eliminating common pre- and postprocessing tasks. The proposed network is trained by using the Ideal Binary Mask (IBM) as the target output label. The IBM identifies the dominant sound source in each T-F bin of the magnitude spectrogram of a mixture signal, by considering each T-F bin as a pixel with a multi-label (for each sound source). Cross entropy is used as the training objective, so as to minimize the average probability error between the target and predicted label for each pixel. By treating the singing voice separation problem as a pixel-wise classification task, we additionally eliminate one of the commonly used, yet not easy to comprehend, postprocessing steps: the Wiener filter postprocessing. The proposed CNN outperforms the first runner up in the Music Information Retrieval Evaluation eXchange (MIREX) 2016 and the winner of MIREX 2014 with a gain of 2.2702 ~ 5.9563 dB global normalized source to distortion ratio (GNSDR) when applied to the iKala dataset. An experiment with the DSD100 dataset on the full-tracks song evaluation task also shows that our model is able to compete with cutting-edge singing voice separation systems which use multi-channel modeling, data augmentation, and model blending.

arXiv:1812.01233공식 코드 · Papers with Code

Spatio-Temporal Action Graph Networks

Events defined by the interaction of objects in a scene are often of critical importance; yet important events may have insufficient labeled examples to train a conventional deep model to generalize to future object appearance. Activity recognition models that represent object interactions explicitly have the potential to learn in a more efficient manner than those that represent scenes with global descriptors. We propose a novel inter-object graph representation for activity recognition based on a disentangled graph embedding with direct observation of edge appearance. We employ a novel factored embedding of the graph structure, disentangling a representation hierarchy formed over spatial dimensions from that found over temporal variation. We demonstrate the effectiveness of our model on the Charades activity recognition benchmark, as well as a new dataset of driving activities focusing on multi-object interactions with near-collision events. Our model offers significantly improved performance compared to baseline approaches without object-graph representations, or with previous graph-based models.

arXiv:1812.01222

Ladder Networks for Semi-Supervised Hyperspectral Image Classification

We used the Ladder Network [Rasmus et al. (2015)] to perform Hyperspectral Image Classification in a semi-supervised setting. The Ladder Network distinguishes itself from other semi-supervised methods by jointly optimizing a supervised and unsupervised cost. In many settings this has proven to be more successful than other semi-supervised techniques, such as pretraining using unlabeled data. We furthermore show that the convolutional Ladder Network outperforms most of the current techniques used in hyperspectral image classification and achieves new state-of-the-art performance on the Pavia University dataset given only 5 labeled data points per class.

arXiv:1812.01216

Parameter Re-Initialization through Cyclical Batch Size Schedules

Optimal parameter initialization remains a crucial problem for neural network training. A poor weight initialization may take longer to train and/or converge to sub-optimal solutions. Here, we propose a method of weight re-initialization by repeated annealing and injection of noise in the training process. We implement this through a cyclical batch size schedule motivated by a Bayesian perspective of neural network training. We evaluate our methods through extensive experiments on tasks in language modeling, natural language inference, and image classification. We demonstrate the ability of our method to improve language modeling performance by up to 7.91 perplexity and reduce training iterations by up to $61\%$, in addition to its flexibility in enabling snapshot ensembling and use with adversarial training.

arXiv:1812.01214

Prototype-based Neural Network Layers: Incorporating Vector Quantization

Neural networks currently dominate the machine learning community and they do so for good reasons. Their accuracy on complex tasks such as image classification is unrivaled at the moment and with recent improvements they are reasonably easy to train. Nevertheless, neural networks are lacking robustness and interpretability. Prototype-based vector quantization methods on the other hand are known for being robust and interpretable. For this reason, we propose techniques and strategies to merge both approaches. This contribution will particularly highlight the similarities between them and outline how to construct a prototype-based classification layer for multilayer networks. Additionally, we provide an alternative, prototype-based, approach to the classical convolution operation. Numerical results are not part of this report, instead the focus lays on establishing a strong theoretical framework. By publishing our framework and the respective theoretical considerations and justifications before finalizing our numerical experiments we hope to jump-start the incorporation of prototype-based learning in neural networks and vice versa.

arXiv:1812.01207

Practical Text Classification With Large Pre-Trained Language Models

Multi-emotion sentiment classification is a natural language processing (NLP) problem with valuable use cases on real-world data. We demonstrate that large-scale unsupervised language modeling combined with finetuning offers a practical solution to this task on difficult datasets, including those with label class imbalance and domain-specific context. By training an attention-based Transformer network (Vaswani et al. 2017) on 40GB of text (Amazon reviews) (McAuley et al. 2015) and fine-tuning on the training set, our model achieves a 0.69 F1 score on the SemEval Task 1:E-c multi-dimensional emotion classification problem (Mohammad et al. 2018), based on the Plutchik wheel of emotions (Plutchik 1979). These results are competitive with state of the art models, including strong F1 scores on difficult (emotion) categories such as Fear (0.73), Disgust (0.77) and Anger (0.78), as well as competitive results on rare categories such as Anticipation (0.42) and Surprise (0.37). Furthermore, we demonstrate our application on a real world text classification task. We create a narrowly collected text dataset of real tweets on several topics, and show that our finetuned model outperforms general purpose commercially available APIs for sentiment and multidimensional emotion classification on this dataset by a significant margin. We also perform a variety of additional studies, investigating properties of deep learning architectures, datasets and algorithms for achieving practical multidimensional sentiment classification. Overall, we find that unsupervised language modeling and finetuning is a simple framework for achieving high quality results on real-world sentiment classification.

arXiv:1812.01187공식 코드 · Papers with Code

Bag of Tricks for Image Classification with Convolutional Neural Networks

Much of the recent progress made in image classification research can be credited to training procedure refinements, such as changes in data augmentations and optimization methods. In the literature, however, most refinements are either briefly mentioned as implementation details or only visible in source code. In this paper, we will examine a collection of such refinements and empirically evaluate their impact on the final model accuracy through ablation study. We will show that, by combining these refinements together, we are able to improve various CNN models significantly. For example, we raise ResNet-50's top-1 validation accuracy from 75.3% to 79.29% on ImageNet. We will also demonstrate that improvement on image classification accuracy leads to better transfer learning performance in other application domains such as object detection and semantic segmentation.

arXiv:1812.01157

Cross-Classification Clustering: An Efficient Multi-Object Tracking Technique for 3-D Instance Segmentation in Connectomics

Pixel-accurate tracking of objects is a key element in many computer vision applications, often solved by iterated individual object tracking or instance segmentation followed by object matching. Here we introduce cross-classification clustering (3C), a technique that simultaneously tracks complex, interrelated objects in an image stack. The key idea in cross-classification is to efficiently turn a clustering problem into a classification problem by running a logarithmic number of independent classifications per image, letting the cross-labeling of these classifications uniquely classify each pixel to the object labels. We apply the 3C mechanism to achieve state-of-the-art accuracy in connectomics -- the nanoscale mapping of neural tissue from electron microscopy volumes. Our reconstruction system increases scalability by an order of magnitude over existing single-object tracking methods (such as flood-filling networks). This scalability is important for the deployment of connectomics pipelines, since currently the best performing techniques require computing infrastructures that are beyond the reach of most laboratories. Our algorithm may offer benefits in other domains that require pixel-accurate tracking of multiple objects, such as segmentation of videos and medical imagery.

arXiv:1812.01124

ORACLE: Optimized Radio clAssification through Convolutional neuraL nEtworks

This paper describes the architecture and performance of ORACLE, an approach for detecting a unique radio from a large pool of bit-similar devices (same hardware, protocol, physical address, MAC ID) using only IQ samples at the physical layer. ORACLE trains a convolutional neural network (CNN) that balances computational time and accuracy, showing 99\% classification accuracy for a 16-node USRP X310 SDR testbed and an external database of $>$100 COTS WiFi devices. Our work makes the following contributions: (i) it studies the hardware-centric features within the transmitter chain that causes IQ sample variations; (ii) for an idealized static channel environment, it proposes a CNN architecture requiring only raw IQ samples accessible at the front-end, without channel estimation or prior knowledge of the communication protocol; (iii) for dynamic channels, it demonstrates a principled method of feedback-driven transmitter-side modifications that uses channel estimation at the receiver to increase differentiability for the CNN classifier. The key innovation here is to intentionally introduce controlled imperfections on the transmitter side through software directives, while minimizing the change in bit error rate. Unlike previous work that imposes constant environmental conditions, ORACLE adopts the `train once deploy anywhere' paradigm with near-perfect device classification accuracy.

arXiv:1812.01068

Machine Friendly Machine Learning: Interpretation of Computed Tomography Without Image Reconstruction

Recent advancements in deep learning for automated image processing and classification have accelerated many new applications for medical image analysis. However, most deep learning applications have been developed using reconstructed, human-interpretable medical images. While image reconstruction from raw sensor data is required for the creation of medical images, the reconstruction process only uses a partial representation of all the data acquired. Here we report the development of a system to directly process raw computed tomography (CT) data in sinogram-space, bypassing the intermediary step of image reconstruction. Two classification tasks were evaluated for their feasibility for sinogram-space machine learning: body region identification and intracranial hemorrhage (ICH) detection. Our proposed SinoNet performed favorably compared to conventional reconstructed image-space-based systems for both tasks, regardless of scanning geometries in terms of projections or detectors. Further, SinoNet performed significantly better when using sparsely sampled sinograms than conventional networks operating in image-space. As a result, sinogram-space algorithms could be used in field settings for binary diagnosis testing, triage, and in clinical settings where low radiation dose is desired. These findings also demonstrate another strength of deep learning where it can analyze and interpret sinograms that are virtually impossible for human experts.

arXiv:1812.01043

Identification and Recognition of Rice Diseases and Pests Using Convolutional Neural Networks

An accurate and timely detection of diseases and pests in rice plants can help farmers in applying timely treatment on the plants and thereby can reduce the economic losses substantially. Recent developments in deep learning based convolutional neural networks (CNN) have greatly improved the image classification accuracy. Being motivated by the success of CNNs in image classification, deep learning based approaches have been developed in this paper for detecting diseases and pests from rice plant images. The contribution of this paper is two fold: (i) State-of-the-art large scale architectures such as VGG16 and InceptionV3 have been adopted and fine tuned for detecting and recognizing rice diseases and pests. Experimental results show the effectiveness of these models with real datasets. (ii) Since large scale architectures are not suitable for mobile devices, a two-stage small CNN architecture has been proposed, and compared with the state-of-the-art memory efficient CNN architectures such as MobileNet, NasNet Mobile and SqueezeNet. Experimental results show that the proposed architecture can achieve the desired accuracy of 93.3\% with a significantly reduced model size (e.g., 99\% less size compared to that of VGG16).

arXiv:1812.00497

Using Multitask Learning to Improve 12-Lead Electrocardiogram Classification

We develop a multi-task convolutional neural network (CNN) to classify multiple diagnoses from 12-lead electrocardiograms (ECGs) using a dataset comprised of over 40,000 ECGs, with labels derived from cardiologist clinical interpretations. Since many clinically important classes can occur in low frequencies, approaches are needed to improve performance on rare classes. We compare the performance of several single-class classifiers on rare classes to a multi-headed classifier across all available classes. We demonstrate that the addition of common classes can significantly improve CNN performance on rarer classes when compared to a model trained on the rarer class in isolation. Using this method, we develop a model with high performance as measured by F1 score on multiple clinically relevant classes compared against the gold-standard cardiologist interpretation.

arXiv:1812.01464

Feasibility of Colon Cancer Detection in Confocal Laser Microscopy Images Using Convolution Neural Networks

Histological evaluation of tissue samples is a typical approach to identify colorectal cancer metastases in the peritoneum. For immediate assessment, reliable and real-time in-vivo imaging would be required. For example, intraoperative confocal laser microscopy has been shown to be suitable for distinguishing organs and also malignant and benign tissue. So far, the analysis is done by human experts. We investigate the feasibility of automatic colon cancer classification from confocal laser microscopy images using deep learning models. We overcome very small dataset sizes through transfer learning with state-of-the-art architectures. We achieve an accuracy of 89.1% for cancer detection in the peritoneum which indicates viability as an intraoperative decision support system.

arXiv:1812.01527

Leveraging Multi-grained Sentiment Lexicon Information for Neural Sequence Models

Neural sequence models have achieved great success in sentence-level sentiment classification. However, some models are exceptionally complex or based on expensive features. Some other models recognize the value of existed linguistic resource but utilize it insufficiently. This paper proposes a novel and general method to incorporate lexicon information, including sentiment lexicons(+/-), negation words and intensifiers. Words are annotated in fine-grained and coarse-grained labels. The proposed method first encodes the fine-grained labels into sentiment embedding and concatenates it with word embedding. Second, the coarse-grained labels are utilized to enhance the attention mechanism to give large weight on sentiment-related words. Experimental results show that our method can increase classification accuracy for neural sequence models on both SST-5 and MR dataset. Specifically, the enhanced Bi-LSTM model can even compare with a Tree-LSTM which uses expensive phrase-level annotations. Further analysis shows that in most cases the lexicon resource can offer the right annotations. Besides, the proposed method is capable of overcoming the effect from inevitably wrong annotations.

arXiv:1812.00155

Learning RoI Transformer for Detecting Oriented Objects in Aerial Images

Object detection in aerial images is an active yet challenging task in computer vision because of the birdview perspective, the highly complex backgrounds, and the variant appearances of objects. Especially when detecting densely packed objects in aerial images, methods relying on horizontal proposals for common object detection often introduce mismatches between the Region of Interests (RoIs) and objects. This leads to the common misalignment between the final object classification confidence and localization accuracy. Although rotated anchors have been used to tackle this problem, the design of them always multiplies the number of anchors and dramatically increases the computational complexity. In this paper, we propose a RoI Transformer to address these problems. More precisely, to improve the quality of region proposals, we first designed a Rotated RoI (RRoI) learner to transform a Horizontal Region of Interest (HRoI) into a Rotated Region of Interest (RRoI). Based on the RRoIs, we then proposed a Rotated Position Sensitive RoI Align (RPS-RoI-Align) module to extract rotation-invariant features from them for boosting subsequent classification and regression. Our RoI Transformer is with light weight and can be easily embedded into detectors for oriented object detection. A simple implementation of the RoI Transformer has achieved state-of-the-art performances on two common and challenging aerial datasets, i.e., DOTA and HRSC2016, with a neglectable reduction to detection speed. Our RoI Transformer exceeds the deformable Position Sensitive RoI pooling when oriented bounding-box annotations are available. Extensive experiments have also validated the flexibility and effectiveness of our RoI Transformer. The results demonstrate that it can be easily integrated with other detector architectures and significantly improve the performances.

arXiv:1811.12649공식 코드 · Papers with Code

Classification is a Strong Baseline for Deep Metric Learning

Deep metric learning aims to learn a function mapping image pixels to embedding feature vectors that model the similarity between images. Two major applications of metric learning are content-based image retrieval and face verification. For the retrieval tasks, the majority of current state-of-the-art (SOTA) approaches are triplet-based non-parametric training. For the face verification tasks, however, recent SOTA approaches have adopted classification-based parametric training. In this paper, we look into the effectiveness of classification based approaches on image retrieval datasets. We evaluate on several standard retrieval datasets such as CAR-196, CUB-200-2011, Stanford Online Product, and In-Shop datasets for image retrieval and clustering, and establish that our classification-based approach is competitive across different feature dimensions and base feature networks. We further provide insights into the performance effects of subsampling classes for scalable classification-based training, and the effects of binarization, enabling efficient storage and computation for practical applications.

arXiv:1811.12611

Virtual Class Enhanced Discriminative Embedding Learning

Recently, learning discriminative features to improve the recognition performances gradually becomes the primary goal of deep learning, and numerous remarkable works have emerged. In this paper, we propose a novel yet extremely simple method \textbf{Virtual Softmax} to enhance the discriminative property of learned features by injecting a dynamic virtual negative class into the original softmax. Injecting virtual class aims to enlarge inter-class margin and compress intra-class distribution by strengthening the decision boundary constraint. Although it seems weird to optimize with this additional virtual class, we show that our method derives from an intuitive and clear motivation, and it indeed encourages the features to be more compact and separable. This paper empirically and experimentally demonstrates the superiority of Virtual Softmax, improving the performances on a variety of object classification and face verification tasks.

arXiv:1811.12587

Restricted Boltzmann Machine with Multivalued Hidden Variables: a model suppressing over-fitting

Generalization is one of the most important issues in machine learning problems. In this study, we consider generalization in restricted Boltzmann machines (RBMs). We propose an RBM with multivalued hidden variables, which is a simple extension of conventional RBMs. We demonstrate that the proposed model is better than the conventional model via numerical experiments for contrastive divergence learning with artificial data and a classification problem with MNIST.

arXiv:1811.12563

Deep Multimodal Learning: An Effective Method for Video Classification

Videos have become ubiquitous on the Internet. And video analysis can provide lots of information for detecting and recognizing objects as well as help people understand human actions and interactions with the real world. However, facing data as huge as TB level, effective methods should be applied. Recurrent neural network (RNN) architecture has wildly been used on many sequential learning problems such as Language Model, Time-Series Analysis, etc. In this paper, we propose some variations of RNN such as stacked bidirectional LSTM/GRU network with attention mechanism to categorize large-scale video data. We also explore different multimodal fusion methods. Our model combines both visual and audio information on both video and frame level and received great result. Ensemble methods are also applied. Because of its multimodal characteristics, we decide to call this method Deep Multimodal Learning(DML). Our DML-based model was trained on Google Cloud and our own server and was tested in a well-known video classification competition on Kaggle held by Google.