Paper Group ANR 498
When CTC Training Meets Acoustic Landmarks. Segmentation of Head and Neck Organs at Risk Using CNN with Batch Dice Loss. A Comparative Study: Adaptive Fuzzy Inference Systems for Energy Prediction in Urban Buildings. Detecting Adversarial Examples via Key-based Network. Bangla License Plate Recognition Using Convolutional Neural Networks (CNN). Bin …
When CTC Training Meets Acoustic Landmarks
Title | When CTC Training Meets Acoustic Landmarks |
Authors | Di He, Xuesong Yang, Boon Pang Lim, Yi Liang, Mark Hasegawa-Johnson, Deming Chen |
Abstract | Connectionist temporal classification (CTC) provides an end-to-end acoustic model (AM) training strategy. CTC learns accurate AMs without time-aligned phonetic transcription, but sometimes fails to converge, especially in resource-constrained scenarios. In this paper, the convergence properties of CTC are improved by incorporating acoustic landmarks. We tailored a new set of acoustic landmarks to help CTC training converge more rapidly and smoothly while also reducing recognition error rates. We leveraged new target label sequences mixed with both phone and manner changes to guide CTC training. Experiments on TIMIT demonstrated that CTC based acoustic models converge significantly faster and smoother when they are augmented by acoustic landmarks. The models pretrained with mixed target labels can be further finetuned, resulting in phone error rates 8.72% below baseline on TIMIT. Consistent performance gain is also observed on WSJ (a larger corpus) and reduced TIMIT (smaller). With WSJ, we are the first to succeed in verifying the effectiveness of acoustic landmark theory on a mid-sized ASR task. |
Tasks | Speech Recognition |
Published | 2018-11-05 |
URL | http://arxiv.org/abs/1811.02063v2 |
http://arxiv.org/pdf/1811.02063v2.pdf | |
PWC | https://paperswithcode.com/paper/when-ctc-training-meets-acoustic-landmarks |
Repo | |
Framework | |
Segmentation of Head and Neck Organs at Risk Using CNN with Batch Dice Loss
Title | Segmentation of Head and Neck Organs at Risk Using CNN with Batch Dice Loss |
Authors | Oldřich Kodym, Michal Španěl, Adam Herout |
Abstract | This paper deals with segmentation of organs at risk (OAR) in head and neck area in CT images which is a crucial step for reliable intensity modulated radiotherapy treatment. We introduce a convolution neural network with encoder-decoder architecture and a new loss function, the batch soft Dice loss function, used to train the network. The resulting model produces segmentations of every OAR in the public MICCAI 2015 Head And Neck Auto-Segmentation Challenge dataset. Despite the heavy class imbalance in the data, we improve accuracy of current state-of-the-art methods by 0.33 mm in terms of average surface distance and by 0.11 in terms of Dice overlap coefficient on average. |
Tasks | |
Published | 2018-12-06 |
URL | http://arxiv.org/abs/1812.02427v1 |
http://arxiv.org/pdf/1812.02427v1.pdf | |
PWC | https://paperswithcode.com/paper/segmentation-of-head-and-neck-organs-at-risk |
Repo | |
Framework | |
A Comparative Study: Adaptive Fuzzy Inference Systems for Energy Prediction in Urban Buildings
Title | A Comparative Study: Adaptive Fuzzy Inference Systems for Energy Prediction in Urban Buildings |
Authors | Mainak Dan, Seshadhri Srinivasan |
Abstract | This investigation aims to study different adaptive fuzzy inference algorithms capable of real-time sequential learning and prediction of time-series data. A brief qualitative description of these algorithms namely meta-cognitive fuzzy inference system (McFIS), sequential adaptive fuzzy inference system (SAFIS) and evolving Takagi-Sugeno (ETS) model provide a comprehensive comparison of their working principle, especially their unique characteristics are discussed. These algorithms are then simulated with dataset collected at one of the academic buildings at Nanyang Technological University, Singapore. The performance are compared by means of the root mean squared error (RMSE) and non-destructive error index (NDEI) of the predicted output. Analysis shows that McFIS shows promising results either with lower RMSE and NDEI or with lower architectural complexity over ETS and SAFIS. Statistical Analysis also reveals the significance of the outcome of these algorithms. |
Tasks | Time Series |
Published | 2018-09-24 |
URL | http://arxiv.org/abs/1809.08860v1 |
http://arxiv.org/pdf/1809.08860v1.pdf | |
PWC | https://paperswithcode.com/paper/a-comparative-study-adaptive-fuzzy-inference |
Repo | |
Framework | |
Detecting Adversarial Examples via Key-based Network
Title | Detecting Adversarial Examples via Key-based Network |
Authors | Pinlong Zhao, Zhouyu Fu, Ou wu, Qinghua Hu, Jun Wang |
Abstract | Though deep neural networks have achieved state-of-the-art performance in visual classification, recent studies have shown that they are all vulnerable to the attack of adversarial examples. Small and often imperceptible perturbations to the input images are sufficient to fool the most powerful deep neural networks. Various defense methods have been proposed to address this issue. However, they either require knowledge on the process of generating adversarial examples, or are not robust against new attacks specifically designed to penetrate the existing defense. In this work, we introduce key-based network, a new detection-based defense mechanism to distinguish adversarial examples from normal ones based on error correcting output codes, using the binary code vectors produced by multiple binary classifiers applied to randomly chosen label-sets as signatures to match normal images and reject adversarial examples. In contrast to existing defense methods, the proposed method does not require knowledge of the process for generating adversarial examples and can be applied to defend against different types of attacks. For the practical black-box and gray-box scenarios, where the attacker does not know the encoding scheme, we show empirically that key-based network can effectively detect adversarial examples generated by several state-of-the-art attacks. |
Tasks | |
Published | 2018-06-02 |
URL | http://arxiv.org/abs/1806.00580v1 |
http://arxiv.org/pdf/1806.00580v1.pdf | |
PWC | https://paperswithcode.com/paper/detecting-adversarial-examples-via-key-based |
Repo | |
Framework | |
Bangla License Plate Recognition Using Convolutional Neural Networks (CNN)
Title | Bangla License Plate Recognition Using Convolutional Neural Networks (CNN) |
Authors | M M Shaifur Rahman, Mst Shamima Nasrin, Moin Mostakim, Md Zahangir Alom |
Abstract | In the last few years, the deep learning technique in particular Convolutional Neural Networks (CNNs) is using massively in the field of computer vision and machine learning. This deep learning technique provides state-of-the-art accuracy in different classification, segmentation, and detection tasks on different benchmarks such as MNIST, CIFAR-10, CIFAR-100, Microsoft COCO, and ImageNet. However, there are a lot of research has been conducted for Bangla License plate recognition with traditional machine learning approaches in last decade. None of them are used to deploy a physical system for Bangla License Plate Recognition System (BLPRS) due to their poor recognition accuracy. In this paper, we have implemented CNNs based Bangla license plate recognition system with better accuracy that can be applied for different purposes including roadside assistance, automatic parking lot management system, vehicle license status detection and so on. Along with that, we have also created and released a very first and standard database for BLPRS. |
Tasks | License Plate Recognition |
Published | 2018-09-04 |
URL | http://arxiv.org/abs/1809.00905v1 |
http://arxiv.org/pdf/1809.00905v1.pdf | |
PWC | https://paperswithcode.com/paper/bangla-license-plate-recognition-using |
Repo | |
Framework | |
Binary Input Layer: Training of CNN models with binary input data
Title | Binary Input Layer: Training of CNN models with binary input data |
Authors | Robert Dürichen, Thomas Rocznik, Oliver Renz, Christian Peters |
Abstract | For the efficient execution of deep convolutional neural networks (CNN) on edge devices, various approaches have been presented which reduce the bit width of the network parameters down to 1 bit. Binarization of the first layer was always excluded, as it leads to a significant error increase. Here, we present the novel concept of binary input layer (BIL), which allows the usage of binary input data by learning bit specific binary weights. The concept is evaluated on three datasets (PAMAP2, SVHN, CIFAR-10). Our results show that this approach is in particular beneficial for multimodal datasets (PAMAP2) where it outperforms networks using full precision weights in the first layer by 1:92 percentage points (pp) while consuming only 2 % of the chip area. |
Tasks | |
Published | 2018-12-09 |
URL | http://arxiv.org/abs/1812.03410v1 |
http://arxiv.org/pdf/1812.03410v1.pdf | |
PWC | https://paperswithcode.com/paper/binary-input-layer-training-of-cnn-models |
Repo | |
Framework | |
Murmur Detection Using Parallel Recurrent & Convolutional Neural Networks
Title | Murmur Detection Using Parallel Recurrent & Convolutional Neural Networks |
Authors | Shahnawaz Alam, Rohan Banerjee, Soma Bandyopadhyay |
Abstract | In this article, we propose a novel technique for classification of the Murmurs in heart sound. We introduce a novel deep neural network architecture using parallel combination of the Recurrent Neural Network (RNN) based Bidirectional Long Short-Term Memory (BiLSTM) & Convolutional Neural Network (CNN) to learn visual and time-dependent characteristics of Murmur in PCG waveform. Set of acoustic features are presented to our proposed deep neural network to discriminate between Normal and Murmur class. The proposed method was evaluated on a large dataset using 5-fold cross-validation, resulting in a sensitivity and specificity of 96 +- 0.6 % , 100 +- 0 % respectively and F1 Score of 98 +- 0.3 %. |
Tasks | |
Published | 2018-08-13 |
URL | http://arxiv.org/abs/1808.04411v1 |
http://arxiv.org/pdf/1808.04411v1.pdf | |
PWC | https://paperswithcode.com/paper/murmur-detection-using-parallel-recurrent |
Repo | |
Framework | |
RIPEx: Extracting malicious IP addresses from security forums using cross-forum learning
Title | RIPEx: Extracting malicious IP addresses from security forums using cross-forum learning |
Authors | Joobin Gharibshah, Evangelos E. Papalexakis, Michalis Faloutsos |
Abstract | Is it possible to extract malicious IP addresses reported in security forums in an automatic way? This is the question at the heart of our work. We focus on security forums, where security professionals and hackers share knowledge and information, and often report misbehaving IP addresses. So far, there have only been a few efforts to extract information from such security forums. We propose RIPEx, a systematic approach to identify and label IP addresses in security forums by utilizing a cross-forum learning method. In more detail, the challenge is twofold: (a) identifying IP addresses from other numerical entities, such as software version numbers, and (b) classifying the IP address as benign or malicious. We propose an integrated solution that tackles both these problems. A novelty of our approach is that it does not require training data for each new forum. Our approach does knowledge transfer across forums: we use a classifier from our source forums to identify seed information for training a classifier on the target forum. We evaluate our method using data collected from five security forums with a total of 31K users and 542K posts. First, RIPEx can distinguish IP address from other numeric expressions with 95% precision and above 93% recall on average. Second, RIPEx identifies malicious IP addresses with an average precision of 88% and over 78% recall, using our cross-forum learning. Our work is a first step towards harnessing the wealth of useful information that can be found in security forums. |
Tasks | Transfer Learning |
Published | 2018-04-13 |
URL | http://arxiv.org/abs/1804.04760v1 |
http://arxiv.org/pdf/1804.04760v1.pdf | |
PWC | https://paperswithcode.com/paper/ripex-extracting-malicious-ip-addresses-from |
Repo | |
Framework | |
Unsupervised Single Image Deraining with Self-supervised Constraints
Title | Unsupervised Single Image Deraining with Self-supervised Constraints |
Authors | Xin Jin, Zhibo Chen, Jianxin Lin, Zhikai Chen, Wei Zhou |
Abstract | Most existing single image deraining methods require learning supervised models from a large set of paired synthetic training data, which limits their generality, scalability and practicality in real-world multimedia applications. Besides, due to lack of labeled-supervised constraints, directly applying existing unsupervised frameworks to the image deraining task will suffer from low-quality recovery. Therefore, we propose an Unsupervised Deraining Generative Adversarial Network (UD-GAN) to tackle above problems by introducing self-supervised constraints from the intrinsic statistics of unpaired rainy and clean images. Specifically, we firstly design two collaboratively optimized modules, namely Rain Guidance Module (RGM) and Background Guidance Module (BGM), to take full advantage of rainy image characteristics: The RGM is designed to discriminate real rainy images from fake rainy images which are created based on outputs of the generator with BGM. Simultaneously, the BGM exploits a hierarchical Gaussian-Blur gradient error to ensure background consistency between rainy input and de-rained output. Secondly, a novel luminance-adjusting adversarial loss is integrated into the clean image discriminator considering the built-in luminance difference between real clean images and derained images. Comprehensive experiment results on various benchmarking datasets and different training settings show that UD-GAN outperforms existing image deraining methods in both quantitative and qualitative comparisons. |
Tasks | Rain Removal, Single Image Deraining |
Published | 2018-11-21 |
URL | http://arxiv.org/abs/1811.08575v1 |
http://arxiv.org/pdf/1811.08575v1.pdf | |
PWC | https://paperswithcode.com/paper/unsupervised-single-image-deraining-with-self |
Repo | |
Framework | |
Loss-Calibrated Approximate Inference in Bayesian Neural Networks
Title | Loss-Calibrated Approximate Inference in Bayesian Neural Networks |
Authors | Adam D. Cobb, Stephen J. Roberts, Yarin Gal |
Abstract | Current approaches in approximate inference for Bayesian neural networks minimise the Kullback-Leibler divergence to approximate the true posterior over the weights. However, this approximation is without knowledge of the final application, and therefore cannot guarantee optimal predictions for a given task. To make more suitable task-specific approximations, we introduce a new loss-calibrated evidence lower bound for Bayesian neural networks in the context of supervised learning, informed by Bayesian decision theory. By introducing a lower bound that depends on a utility function, we ensure that our approximation achieves higher utility than traditional methods for applications that have asymmetric utility functions. Furthermore, in using dropout inference, we highlight that our new objective is identical to that of standard dropout neural networks, with an additional utility-dependent penalty term. We demonstrate our new loss-calibrated model with an illustrative medical example and a restricted model capacity experiment, and highlight failure modes of the comparable weighted cross entropy approach. Lastly, we demonstrate the scalability of our method to real world applications with per-pixel semantic segmentation on an autonomous driving data set. |
Tasks | Autonomous Driving, Semantic Segmentation |
Published | 2018-05-10 |
URL | http://arxiv.org/abs/1805.03901v1 |
http://arxiv.org/pdf/1805.03901v1.pdf | |
PWC | https://paperswithcode.com/paper/loss-calibrated-approximate-inference-in |
Repo | |
Framework | |
Semi-Supervised Confidence Network aided Gated Attention based Recurrent Neural Network for Clickbait Detection
Title | Semi-Supervised Confidence Network aided Gated Attention based Recurrent Neural Network for Clickbait Detection |
Authors | Amrith Rajagopal Setlur |
Abstract | Clickbaits are catchy headlines that are frequently used by social media outlets in order to allure its viewers into clicking them and thus leading them to dubious content. Such venal schemes thrive on exploiting the curiosity of naive social media users, directing traffic to web pages that won’t be visited otherwise. In this paper, we propose a novel, semi-supervised classification based approach, that employs attentions sampled from a Gumbel-Softmax distribution to distill contexts that are fairly important in clickbait detection. An additional loss over the attention weights is used to encode prior knowledge. Furthermore, we propose a confidence network that enables learning over weak labels and improves robustness to noisy labels. We show that with merely 30% of strongly labeled samples we can achieve over 97% of the accuracy, of current state of the art methods in clickbait detection. |
Tasks | Clickbait Detection |
Published | 2018-11-04 |
URL | http://arxiv.org/abs/1811.01355v1 |
http://arxiv.org/pdf/1811.01355v1.pdf | |
PWC | https://paperswithcode.com/paper/semi-supervised-confidence-network-aided |
Repo | |
Framework | |
Uncharted Forest a Technique for Exploratory Data Analysis
Title | Uncharted Forest a Technique for Exploratory Data Analysis |
Authors | Casey Kneale, Steven D. Brown |
Abstract | Exploratory data analysis is crucial for developing and understanding classification models from high-dimensional datasets. We explore the utility of a new unsupervised tree ensemble called uncharted forest for visualizing class associations, sample-sample associations, class heterogeneity, and uninformative classes for provenance studies. The uncharted forest algorithm can be used to partition data using random selections of variables and metrics based on statistical spread. After each tree is grown, a tally of the samples that arrive at every terminal node is maintained. Those tallies are stored in single sample association matrix and a likelihood measure for each sample being partitioned with one another can be made. That matrix may be readily viewed as a heat map, and the probabilities can be quantified via new metrics that account for class or cluster membership. We display the advantages and limitations of using this technique by applying it to two classification datasets and three provenance study datasets. Two of the metrics presented in this paper are also compared with widely used metrics from two algorithms that have variance-based clustering mechanisms. |
Tasks | |
Published | 2018-02-11 |
URL | http://arxiv.org/abs/1802.03840v3 |
http://arxiv.org/pdf/1802.03840v3.pdf | |
PWC | https://paperswithcode.com/paper/uncharted-forest-a-technique-for-exploratory |
Repo | |
Framework | |
Component-based Attention for Large-scale Trademark Retrieval
Title | Component-based Attention for Large-scale Trademark Retrieval |
Authors | Osman Tursun, Simon Denman, Sabesan Sivapalan, Sridha Sridharan, Clinton Fookes, Sandra Mau |
Abstract | The demand for large-scale trademark retrieval (TR) systems has significantly increased to combat the rise in international trademark infringement. Unfortunately, the ranking accuracy of current approaches using either hand-crafted or pre-trained deep convolution neural network (DCNN) features is inadequate for large-scale deployments. We show in this paper that the ranking accuracy of TR systems can be significantly improved by incorporating hard and soft attention mechanisms, which direct attention to critical information such as figurative elements and reduce attention given to distracting and uninformative elements such as text and background. Our proposed approach achieves state-of-the-art results on a challenging large-scale trademark dataset. |
Tasks | Trademark Retrieval |
Published | 2018-11-07 |
URL | https://arxiv.org/abs/1811.02746v2 |
https://arxiv.org/pdf/1811.02746v2.pdf | |
PWC | https://paperswithcode.com/paper/component-based-attention-for-large-scale |
Repo | |
Framework | |
Automatic Normalization of Word Variations in Code-Mixed Social Media Text
Title | Automatic Normalization of Word Variations in Code-Mixed Social Media Text |
Authors | Rajat Singh, Nurendra Choudhary, Manish Shrivastava |
Abstract | Social media platforms such as Twitter and Facebook are becoming popular in multilingual societies. This trend induces portmanteau of South Asian languages with English. The blend of multiple languages as code-mixed data has recently become popular in research communities for various NLP tasks. Code-mixed data consist of anomalies such as grammatical errors and spelling variations. In this paper, we leverage the contextual property of words where the different spelling variation of words share similar context in a large noisy social media text. We capture different variations of words belonging to same context in an unsupervised manner using distributed representations of words. Our experiments reveal that preprocessing of the code-mixed dataset based on our approach improves the performance in state-of-the-art part-of-speech tagging (POS-tagging) and sentiment analysis tasks. |
Tasks | Part-Of-Speech Tagging, Sentiment Analysis |
Published | 2018-04-03 |
URL | http://arxiv.org/abs/1804.00804v1 |
http://arxiv.org/pdf/1804.00804v1.pdf | |
PWC | https://paperswithcode.com/paper/automatic-normalization-of-word-variations-in |
Repo | |
Framework | |
Anomaly Detection via Minimum Likelihood Generative Adversarial Networks
Title | Anomaly Detection via Minimum Likelihood Generative Adversarial Networks |
Authors | Chu Wang, Yan-Ming Zhang, Cheng-Lin Liu |
Abstract | Anomaly detection aims to detect abnormal events by a model of normality. It plays an important role in many domains such as network intrusion detection, criminal activity identity and so on. With the rapidly growing size of accessible training data and high computation capacities, deep learning based anomaly detection has become more and more popular. In this paper, a new domain-based anomaly detection method based on generative adversarial networks (GAN) is proposed. Minimum likelihood regularization is proposed to make the generator produce more anomalies and prevent it from converging to normal data distribution. Proper ensemble of anomaly scores is shown to improve the stability of discriminator effectively. The proposed method has achieved significant improvement than other anomaly detection methods on Cifar10 and UCI datasets. |
Tasks | Anomaly Detection, Intrusion Detection, Network Intrusion Detection |
Published | 2018-08-01 |
URL | http://arxiv.org/abs/1808.00200v1 |
http://arxiv.org/pdf/1808.00200v1.pdf | |
PWC | https://paperswithcode.com/paper/anomaly-detection-via-minimum-likelihood |
Repo | |
Framework | |