October 17, 2019

2987 words 15 mins read

Paper Group ANR 970

Detecting Hate Speech and Offensive Language on Twitter using Machine Learning: An N-gram and TFIDF based Approach. An Imprecise Probabilistic Estimator for the Transition Rate Matrix of a Continuous-Time Markov Chain. Adapting End-to-End Neural Speaker Verification to New Languages and Recording Conditions with Adversarial Training. Adversarial Tr …

Detecting Hate Speech and Offensive Language on Twitter using Machine Learning: An N-gram and TFIDF based Approach


Title	Detecting Hate Speech and Offensive Language on Twitter using Machine Learning: An N-gram and TFIDF based Approach
Authors	Aditya Gaydhani, Vikrant Doma, Shrikant Kendre, Laxmi Bhagwat
Abstract	Toxic online content has become a major issue in today’s world due to an exponential increase in the use of internet by people of different cultures and educational background. Differentiating hate speech and offensive language is a key challenge in automatic detection of toxic text content. In this paper, we propose an approach to automatically classify tweets on Twitter into three classes: hateful, offensive and clean. Using Twitter dataset, we perform experiments considering n-grams as features and passing their term frequency-inverse document frequency (TFIDF) values to multiple machine learning models. We perform comparative analysis of the models considering several values of n in n-grams and TFIDF normalization methods. After tuning the model giving the best results, we achieve 95.6% accuracy upon evaluating it on test data. We also create a module which serves as an intermediate between user and Twitter.
Tasks
Published	2018-09-23
URL	http://arxiv.org/abs/1809.08651v1
PDF	http://arxiv.org/pdf/1809.08651v1.pdf
PWC	https://paperswithcode.com/paper/detecting-hate-speech-and-offensive-language
Repo
Framework

An Imprecise Probabilistic Estimator for the Transition Rate Matrix of a Continuous-Time Markov Chain


Title	An Imprecise Probabilistic Estimator for the Transition Rate Matrix of a Continuous-Time Markov Chain
Authors	Thomas Krak, Alexander Erreygers, Jasper De Bock
Abstract	We consider the problem of estimating the transition rate matrix of a continuous-time Markov chain from a finite-duration realisation of this process. We approach this problem in an imprecise probabilistic framework, using a set of prior distributions on the unknown transition rate matrix. The resulting estimator is a set of transition rate matrices that, for reasons of conjugacy, is easy to find. To determine the hyperparameters for our set of priors, we reconsider the problem in discrete time, where we can use the well-known Imprecise Dirichlet Model. In particular, we show how the limit of the resulting discrete-time estimators is a continuous-time estimator. It corresponds to a specific choice of hyperparameters and has an exceptionally simple closed-form expression.
Tasks
Published	2018-04-04
URL	http://arxiv.org/abs/1804.01330v2
PDF	http://arxiv.org/pdf/1804.01330v2.pdf
PWC	https://paperswithcode.com/paper/an-imprecise-probabilistic-estimator-for-the
Repo
Framework

Adapting End-to-End Neural Speaker Verification to New Languages and Recording Conditions with Adversarial Training


Title	Adapting End-to-End Neural Speaker Verification to New Languages and Recording Conditions with Adversarial Training
Authors	Gautam Bhattacharya, Jahangir Alam, Patrick Kenny
Abstract	In this article we propose a novel approach for adapting speaker embeddings to new domains based on adversarial training of neural networks. We apply our embeddings to the task of text-independent speaker verification, a challenging, real-world problem in biometric security. We further the development of end-to-end speaker embedding models by combing a novel 1-dimensional, self-attentive residual network, an angular margin loss function and adversarial training strategy. Our model is able to learn extremely compact, 64-dimensional speaker embeddings that deliver competitive performance on a number of popular datasets using simple cosine distance scoring. One the NIST-SRE 2016 task we are able to beat a strong i-vector baseline, while on the Speakers in the Wild task our model was able to outperform both i-vector and x-vector baselines, showing an absolute improvement of 2.19% over the latter. Additionally, we show that the integration of adversarial training consistently leads to a significant improvement over an unadapted model.
Tasks	Speaker Verification, Text-Independent Speaker Verification
Published	2018-11-07
URL	http://arxiv.org/abs/1811.03055v1
PDF	http://arxiv.org/pdf/1811.03055v1.pdf
PWC	https://paperswithcode.com/paper/adapting-end-to-end-neural-speaker
Repo
Framework

Adversarial Training for Patient-Independent Feature Learning with IVOCT Data for Plaque Classification


Title	Adversarial Training for Patient-Independent Feature Learning with IVOCT Data for Plaque Classification
Authors	Nils Gessert, Markus Heyder, Sarah Latus, David M. Leistner, Youssef S. Abdelwahed, Matthias Lutz, Alexander Schlaefer
Abstract	Deep learning methods have shown impressive results for a variety of medical problems over the last few years. However, datasets tend to be small due to time-consuming annotation. As datasets with different patients are often very heterogeneous generalization to new patients can be difficult. This is complicated further if large differences in image acquisition can occur, which is common during intravascular optical coherence tomography for coronary plaque imaging. We address this problem with an adversarial training strategy where we force a part of a deep neural network to learn features that are independent of patient- or acquisitionspecific characteristics. We compare our regularization method to typical data augmentation strategies and show that our approach improves performance for a small medical dataset.
Tasks	Data Augmentation
Published	2018-05-16
URL	http://arxiv.org/abs/1805.06223v1
PDF	http://arxiv.org/pdf/1805.06223v1.pdf
PWC	https://paperswithcode.com/paper/adversarial-training-for-patient-independent
Repo
Framework

Building medical image classifiers with very limited data using segmentation networks


Title	Building medical image classifiers with very limited data using segmentation networks
Authors	Ken C. L. Wong, Tanveer Syeda-Mahmood, Mehdi Moradi
Abstract	Deep learning has shown promising results in medical image analysis, however, the lack of very large annotated datasets confines its full potential. Although transfer learning with ImageNet pre-trained classification models can alleviate the problem, constrained image sizes and model complexities can lead to unnecessary increase in computational cost and decrease in performance. As many common morphological features are usually shared by different classification tasks of an organ, it is greatly beneficial if we can extract such features to improve classification with limited samples. Therefore, inspired by the idea of curriculum learning, we propose a strategy for building medical image classifiers using features from segmentation networks. By using a segmentation network pre-trained on similar data as the classification task, the machine can first learn the simpler shape and structural concepts before tackling the actual classification problem which usually involves more complicated concepts. Using our proposed framework on a 3D three-class brain tumor type classification problem, we achieved 82% accuracy on 191 testing samples with 91 training samples. When applying to a 2D nine-class cardiac semantic level classification problem, we achieved 86% accuracy on 263 testing samples with 108 training samples. Comparisons with ImageNet pre-trained classifiers and classifiers trained from scratch are presented.
Tasks	Transfer Learning
Published	2018-08-15
URL	http://arxiv.org/abs/1808.05205v1
PDF	http://arxiv.org/pdf/1808.05205v1.pdf
PWC	https://paperswithcode.com/paper/building-medical-image-classifiers-with-very
Repo
Framework

An Analysis of Hierarchical Text Classification Using Word Embeddings


Title	An Analysis of Hierarchical Text Classification Using Word Embeddings
Authors	Roger A. Stein, Patricia A. Jaques, Joao F. Valiati
Abstract	Efficient distributed numerical word representation models (word embeddings) combined with modern machine learning algorithms have recently yielded considerable improvement on automatic document classification tasks. However, the effectiveness of such techniques has not been assessed for the hierarchical text classification (HTC) yet. This study investigates the application of those models and algorithms on this specific problem by means of experimentation and analysis. We trained classification models with prominent machine learning algorithm implementations—fastText, XGBoost, SVM, and Keras’ CNN—and noticeable word embeddings generation methods—GloVe, word2vec, and fastText—with publicly available data and evaluated them with measures specifically appropriate for the hierarchical context. FastText achieved an ${}_{LCA}F_1$ of 0.893 on a single-labeled version of the RCV1 dataset. An analysis indicates that using word embeddings and its flavors is a very promising approach for HTC.
Tasks	Document Classification, Text Classification, Word Embeddings
Published	2018-09-06
URL	http://arxiv.org/abs/1809.01771v1
PDF	http://arxiv.org/pdf/1809.01771v1.pdf
PWC	https://paperswithcode.com/paper/an-analysis-of-hierarchical-text
Repo
Framework

Resource aware design of a deep convolutional-recurrent neural network for speech recognition through audio-visual sensor fusion


Title	Resource aware design of a deep convolutional-recurrent neural network for speech recognition through audio-visual sensor fusion
Authors	Matthijs Van keirsbilck, Bert Moons, Marian Verhelst
Abstract	Today’s Automatic Speech Recognition systems only rely on acoustic signals and often don’t perform well under noisy conditions. Performing multi-modal speech recognition - processing acoustic speech signals and lip-reading video simultaneously - significantly enhances the performance of such systems, especially in noisy environments. This work presents the design of such an audio-visual system for Automated Speech Recognition, taking memory and computation requirements into account. First, a Long-Short-Term-Memory neural network for acoustic speech recognition is designed. Second, Convolutional Neural Networks are used to model lip-reading features. These are combined with an LSTM network to model temporal dependencies and perform automatic lip-reading on video. Finally, acoustic-speech and visual lip-reading networks are combined to process acoustic and visual features simultaneously. An attention mechanism ensures performance of the model in noisy environments. This system is evaluated on the TCD-TIMIT ‘lipspeaker’ dataset for audio-visual phoneme recognition with clean audio and with additive white noise at an SNR of 0dB. It achieves 75.70% and 58.55% phoneme accuracy respectively, over 14 percentage points better than the state-of-the-art for all noise levels.
Tasks	Sensor Fusion, Speech Recognition
Published	2018-03-13
URL	http://arxiv.org/abs/1803.04840v1
PDF	http://arxiv.org/pdf/1803.04840v1.pdf
PWC	https://paperswithcode.com/paper/resource-aware-design-of-a-deep-convolutional
Repo
Framework

On Asynchronous Non-Dominated Sorting for Steady-State Multiobjective Evolutionary Algorithms


Title	On Asynchronous Non-Dominated Sorting for Steady-State Multiobjective Evolutionary Algorithms
Authors	Ilya Yakupov, Maxim Buzdalov
Abstract	In parallel and distributed environments, generational evolutionary algorithms often do not exploit the full potential of the computation system since they have to wait until the entire population is evaluated before starting selection procedures. Steady-state algorithms are often seen as a solution to this problem, since fitness evaluation can be done by multiple threads in an asynchronous way. However, if the algorithm updates its state in a complicated way, the threads will eventually have to wait until this update finishes. State update procedures that are computationally expensive are common in multiobjective evolutionary algorithms. We have implemented an asynchronous steady-state version of the NSGA-II algorithm. Its most expensive part, non-dominated sorting, determines the time needed to update the state. We turned the existing incremental non-dominated sorting algorithm into an asynchronous one using several concurrency techniques: a single entry-level lock, finer-grained locks working with non-domination levels, and a non-blocking approach using compare-and-set operations. Our experimental results reveal the trade-off between the work-efficiency of the algorithm and the achieved amount of parallelism.
Tasks
Published	2018-04-14
URL	http://arxiv.org/abs/1804.05208v1
PDF	http://arxiv.org/pdf/1804.05208v1.pdf
PWC	https://paperswithcode.com/paper/on-asynchronous-non-dominated-sorting-for
Repo
Framework

Reinforcement Learning for Heterogeneous Teams with PALO Bounds


Title	Reinforcement Learning for Heterogeneous Teams with PALO Bounds
Authors	Roi Ceren, Prashant Doshi, Keyang He
Abstract	We introduce reinforcement learning for heterogeneous teams in which rewards for an agent are additively factored into local costs, stimuli unique to each agent, and global rewards, those shared by all agents in the domain. Motivating domains include coordination of varied robotic platforms, which incur different costs for the same action, but share an overall goal. We present two templates for learning in this setting with factored rewards: a generalization of Perkins’ Monte Carlo exploring starts for POMDPs to canonical MPOMDPs, with a single policy mapping joint observations of all agents to joint actions (MCES-MP); and another with each agent individually mapping joint observations to their own action (MCES-FMP). We use probably approximately local optimal (PALO) bounds to analyze sample complexity, instantiating these templates to PALO learning. We promote sample efficiency by including a policy space pruning technique, and evaluate the approaches on three domains of heterogeneous agents demonstrating that MCES-FMP yields improved policies in less samples compared to MCES-MP and a previous benchmark.
Tasks
Published	2018-05-23
URL	http://arxiv.org/abs/1805.09267v1
PDF	http://arxiv.org/pdf/1805.09267v1.pdf
PWC	https://paperswithcode.com/paper/reinforcement-learning-for-heterogeneous
Repo
Framework

DDRprog: A CLEVR Differentiable Dynamic Reasoning Programmer


Title	DDRprog: A CLEVR Differentiable Dynamic Reasoning Programmer
Authors	Joseph Suarez, Justin Johnson, Fei-Fei Li
Abstract	We present a novel Dynamic Differentiable Reasoning (DDR) framework for jointly learning branching programs and the functions composing them; this resolves a significant nondifferentiability inhibiting recent dynamic architectures. We apply our framework to two settings in two highly compact and data efficient architectures: DDRprog for CLEVR Visual Question Answering and DDRstack for reverse Polish notation expression evaluation. DDRprog uses a recurrent controller to jointly predict and execute modular neural programs that directly correspond to the underlying question logic; it explicitly forks subprocesses to handle logical branching. By effectively leveraging additional structural supervision, we achieve a large improvement over previous approaches in subtask consistency and a small improvement in overall accuracy. We further demonstrate the benefits of structural supervision in the RPN setting: the inclusion of a stack assumption in DDRstack allows our approach to generalize to long expressions where an LSTM fails the task.
Tasks	Question Answering, Visual Question Answering
Published	2018-03-30
URL	http://arxiv.org/abs/1803.11361v1
PDF	http://arxiv.org/pdf/1803.11361v1.pdf
PWC	https://paperswithcode.com/paper/ddrprog-a-clevr-differentiable-dynamic
Repo
Framework

Benchmarking five global optimization approaches for nano-optical shape optimization and parameter reconstruction


Title	Benchmarking five global optimization approaches for nano-optical shape optimization and parameter reconstruction
Authors	Philipp-Immanuel Schneider, Xavier Garcia Santiago, Victor Soltwisch, Martin Hammerschmidt, Sven Burger, Carsten Rockstuhl
Abstract	Numerical optimization is an important tool in the field of computational physics in general and in nano-optics in specific. It has attracted attention with the increase in complexity of structures that can be realized with nowadays nano-fabrication technologies for which a rational design is no longer feasible. Also, numerical resources are available to enable the computational photonic material design and to identify structures that meet predefined optical properties for specific applications. However, the optimization objective function is in general non-convex and its computation remains resource demanding such that the right choice for the optimization method is crucial to obtain excellent results. Here, we benchmark five global optimization methods for three typical nano-optical optimization problems: \removed{downhill simplex optimization, the limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) algorithm, particle swarm optimization, differential evolution, and Bayesian optimization} \added{particle swarm optimization, differential evolution, and Bayesian optimization as well as multi-start versions of downhill simplex optimization and the limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) algorithm}. In the shown examples from the field of shape optimization and parameter reconstruction, Bayesian optimization, mainly known from machine learning applications, obtains significantly better results in a fraction of the run times of the other optimization methods.
Tasks
Published	2018-09-18
URL	https://arxiv.org/abs/1809.06674v3
PDF	https://arxiv.org/pdf/1809.06674v3.pdf
PWC	https://paperswithcode.com/paper/benchmarking-five-global-optimization
Repo
Framework

The Multiscale Bowler-Hat Transform for Vessel Enhancement in 3D Biomedical Images


Title	The Multiscale Bowler-Hat Transform for Vessel Enhancement in 3D Biomedical Images
Authors	Cigdem Sazak, Carl J. Nelson, Boguslaw Obara
Abstract	Enhancement and detection of 3D vessel-like structures has long been an open problem as most existing image processing methods fail in many aspects, including a lack of uniform enhancement between vessels of different radii and a lack of enhancement at the junctions. Here, we propose a method based on mathematical morphology to enhance 3D vessel-like structures in biomedical images. The proposed method, 3D bowler-hat transform, combines sphere and line structuring elements to enhance vessel-like structures. The proposed method is validated on synthetic and real data and compared with state-of-the-art methods. Our results show that the proposed method achieves a high-quality vessel-like structures enhancement in both synthetic and real biomedical images, and is able to cope with variations in vessels thickness throughout vascular networks while remaining robust at junctions.
Tasks
Published	2018-02-14
URL	http://arxiv.org/abs/1802.05097v1
PDF	http://arxiv.org/pdf/1802.05097v1.pdf
PWC	https://paperswithcode.com/paper/the-multiscale-bowler-hat-transform-for
Repo
Framework

Fingerprint Match in Box


Title	Fingerprint Match in Box
Authors	Joshua J. Engelsma, Kai Cao, Anil K. Jain
Abstract	We open source fingerprint Match in Box, a complete end-to-end fingerprint recognition system embedded within a 4 inch cube. Match in Box stands in contrast to a typical bulky and expensive proprietary fingerprint recognition system which requires sending a fingerprint image to an external host for processing and subsequent spoof detection and matching. In particular, Match in Box is a first of a kind, portable, low-cost, and easy-to-assemble fingerprint reader with an enrollment database embedded within the reader’s memory and open source fingerprint spoof detector, feature extractor, and matcher all running on the reader’s internal vision processing unit (VPU). An onboard touch screen and rechargeable battery pack make this device extremely portable and ideal for applying both fingerprint authentication (1:1 comparison) and fingerprint identification (1:N search) to applications (vaccination tracking, food and benefit distribution programs, human trafficking prevention) in rural communities, especially in developing countries. We also show that Match in Box is suited for capturing neonate fingerprints due to its high resolution (1900 ppi) cameras.
Tasks
Published	2018-04-23
URL	http://arxiv.org/abs/1804.08659v1
PDF	http://arxiv.org/pdf/1804.08659v1.pdf
PWC	https://paperswithcode.com/paper/fingerprint-match-in-box
Repo
Framework

ParaNet - Using Dense Blocks for Early Inference


Title	ParaNet - Using Dense Blocks for Early Inference
Authors	Joseph Chuang, Eric Tsai, Kevin Huang, Jay Fetter
Abstract	DenseNets have been shown to be a competitive model among recent convolutional network architectures. These networks utilize Dense Blocks, which are groups of densely connected layers where the output of a hidden layer is fed in as the input of every other layer following it. In this paper, we aim to improve certain aspects of DenseNet, especially when it comes to practicality. We introduce ParaNet, a new architecture that constructs three pipelines which allow for early inference. We additionally introduce a cascading mechanism such that different pipelines are able to share parameters, as well as logit matching between the outputs of the pipelines. We separately evaluate each of the newly introduced mechanisms of ParaNet, then evaluate our proposed architecture on CIFAR-100.
Tasks
Published	2018-08-24
URL	http://arxiv.org/abs/1808.08308v1
PDF	http://arxiv.org/pdf/1808.08308v1.pdf
PWC	https://paperswithcode.com/paper/paranet-using-dense-blocks-for-early
Repo
Framework

Modeling Task Effects in Human Reading with Neural Attention


Title	Modeling Task Effects in Human Reading with Neural Attention
Authors	Michael Hahn, Frank Keller
Abstract	Humans read by making a sequence of fixations and saccades. They often skip words, without apparent detriment to understanding. We offer a novel explanation for skipping: readers optimize a tradeoff between performing a language-related task and fixating as few words as possible. We propose a neural architecture that combines an attention module (deciding whether to skip words) and a task module (memorizing the input). We show that our model predicts human skipping behavior, while also modeling reading times well, even though it skips 40% of the input. A key prediction of our model is that different reading tasks should result in different skipping behaviors. We confirm this prediction in an eye-tracking experiment in which participants answers questions about a text. We are able to capture these experimental results using the our model, replacing the memorization module with a task module that performs neural question answering.
Tasks	Eye Tracking, Question Answering
Published	2018-07-31
URL	http://arxiv.org/abs/1808.00054v2
PDF	http://arxiv.org/pdf/1808.00054v2.pdf
PWC	https://paperswithcode.com/paper/modeling-task-effects-in-human-reading-with
Repo
Framework