January 28, 2020

3556 words 17 mins read

Paper Group ANR 905

CMB-GAN: Fast Simulations of Cosmic Microwave background anisotropy maps using Deep Learning. Headline Generation: Learning from Decomposable Document Titles. DLIMD: Dictionary Learning based Image-domain Material Decomposition for spectral CT. Learning to Fix Build Errors with Graph2Diff Neural Networks. Ultra-Reliable and Low-Latency Vehicular Co …

CMB-GAN: Fast Simulations of Cosmic Microwave background anisotropy maps using Deep Learning


Title	CMB-GAN: Fast Simulations of Cosmic Microwave background anisotropy maps using Deep Learning
Authors	Amit Mishra, Pranath Reddy, Rahul Nigam
Abstract	Cosmic Microwave Background (CMB) has been a cornerstone in many cosmology experiments and studies since it was discovered back in 1964. Traditional computational models like CAMB that are used for generating CMB temperature anisotropy maps are extremely resource intensive and act as a bottleneck in cosmology experiments that require a large amount of CMB data for analysis. In this paper, we present a new approach to the generation of CMB temperature maps using a specific class of neural networks called Generative Adversarial Network (GAN). We train our deep generative model to learn the complex distribution of CMB maps and efficiently generate new sets of CMB data in the form of 2D patches of anisotropy maps without losing much accuracy. We limit our experiment to the generation of 56$^{\circ}$ and 112$^{\circ}$ square patches of CMB maps. We have also trained a Multilayer perceptron model for estimation of baryon density from a CMB map, we will be using this model for the performance evaluation of our generative model using diagnostic measures like Histogram of pixel intensities, the standard deviation of pixel intensity distribution, Power Spectrum, Cross power spectrum, Correlation matrix of the power spectrum and Peak count. We show that the GAN model is able to efficiently generate CMB samples of multiple sizes and is sensitive to the cosmological parameters corresponding to the underlying distribution of the data. The primiary advantage of this method is the exponential reduction in the computational time needed to generate the CMB data, the GAN model is able to generate the samples within seconds as opposed to hours required by the CAMB package with an acceptable value to error and loss of information. We hope that future iterations of this methodology will replace traditional statistical methods of CMB data generation and help in large scale cosmological experiments.
Tasks
Published	2019-08-11
URL	https://arxiv.org/abs/1908.04682v3
PDF	https://arxiv.org/pdf/1908.04682v3.pdf
PWC	https://paperswithcode.com/paper/cmb-gan-fast-simulations-of-cosmic-microwave
Repo
Framework

Headline Generation: Learning from Decomposable Document Titles


Title	Headline Generation: Learning from Decomposable Document Titles
Authors	Oleg Vasilyev, Tom Grek, John Bohannon
Abstract	We propose a novel method for generating titles for unstructured text documents. We reframe the problem as a sequential question-answering task. A deep neural network is trained on document-title pairs with decomposable titles, meaning that the vocabulary of the title is a subset of the vocabulary of the document. To train the model we use a corpus of millions of publicly available document-title pairs: news articles and headlines. We present the results of a randomized double-blind trial in which subjects were unaware of which titles were human or machine-generated. When trained on approximately 1.5 million news articles, the model generates headlines that humans judge to be as good or better than the original human-written headlines in the majority of cases.
Tasks	Question Answering
Published	2019-04-17
URL	https://arxiv.org/abs/1904.08455v3
PDF	https://arxiv.org/pdf/1904.08455v3.pdf
PWC	https://paperswithcode.com/paper/headline-generation-learning-from-decomposed
Repo
Framework

DLIMD: Dictionary Learning based Image-domain Material Decomposition for spectral CT


Title	DLIMD: Dictionary Learning based Image-domain Material Decomposition for spectral CT
Authors	Weiwen Wu, Haijun Yu, Peijun Chen, Fulin Luo, Fenglin Liu, Qian Wang, Yining Zhu, Yanbo Zhang, Jian Feng, Hengyong Yu
Abstract	The potential huge advantage of spectral computed tomography (CT) is its capability to provide accuracy material identification and quantitative tissue information. This can benefit clinical applications, such as brain angiography, early tumor recognition, etc. To achieve more accurate material components with higher material image quality, we develop a dictionary learning based image-domain material decomposition (DLIMD) for spectral CT in this paper. First, we reconstruct spectral CT image from projections and calculate material coefficients matrix by selecting uniform regions of basis materials from image reconstruction results. Second, we employ the direct inversion (DI) method to obtain initial material decomposition results, and a set of image patches are extracted from the mode-1 unfolding of normalized material image tensor to train a united dictionary by the K-SVD technique. Third, the trained dictionary is employed to explore the similarities from decomposed material images by constructing the DLIMD model. Fourth, more constraints (i.e., volume conservation and the bounds of each pixel within material maps) are further integrated into the model to improve the accuracy of material decomposition. Finally, both physical phantom and preclinical experiments are employed to evaluate the performance of the proposed DLIMD in material decomposition accuracy, material image edge preservation and feature recovery.
Tasks	Computed Tomography (CT), Dictionary Learning, Image Reconstruction
Published	2019-05-06
URL	https://arxiv.org/abs/1905.02567v2
PDF	https://arxiv.org/pdf/1905.02567v2.pdf
PWC	https://paperswithcode.com/paper/dlimd-dictionary-learning-based-image-domain
Repo
Framework

Learning to Fix Build Errors with Graph2Diff Neural Networks


Title	Learning to Fix Build Errors with Graph2Diff Neural Networks
Authors	Daniel Tarlow, Subhodeep Moitra, Andrew Rice, Zimin Chen, Pierre-Antoine Manzagol, Charles Sutton, Edward Aftandilian
Abstract	Professional software developers spend a significant amount of time fixing builds, but this has received little attention as a problem in automatic program repair. We present a new deep learning architecture, called Graph2Diff, for automatically localizing and fixing build errors. We represent source code, build configuration files, and compiler diagnostic messages as a graph, and then use a Graph Neural Network model to predict a diff. A diff specifies how to modify the code’s abstract syntax tree, represented in the neural network as a sequence of tokens and of pointers to code locations. Our network is an instance of a more general abstraction that we call Graph2Tocopo, which is potentially useful in any development tool for predicting source code changes. We evaluate the model on a dataset of over 500k real build errors and their resolutions from professional developers. Compared to the approach of DeepDelta (Mesbah et al., 2019), our approach tackles the harder task of predicting a more precise diff but still achieves over double the accuracy.
Tasks
Published	2019-11-04
URL	https://arxiv.org/abs/1911.01205v1
PDF	https://arxiv.org/pdf/1911.01205v1.pdf
PWC	https://paperswithcode.com/paper/learning-to-fix-build-errors-with-graph2diff
Repo
Framework

Ultra-Reliable and Low-Latency Vehicular Communication: An Active Learning Approach


Title	Ultra-Reliable and Low-Latency Vehicular Communication: An Active Learning Approach
Authors	Mohamed K. Abdel-Aziz, Sumudu Samarakoon, Mehdi Bennis, Walid Saad
Abstract	In this letter, an age of information (AoI)-aware transmission power and resource block (RB) allocation technique for vehicular communication networks is proposed. Due to the highly dynamic nature of vehicular networks, gaining a prior knowledge about the network dynamics, i.e., wireless channels and interference, in order to allocate resources, is challenging. Therefore, to effectively allocate power and RBs, the proposed approach allows the network to actively learn its dynamics by balancing a tradeoff between minimizing the probability that the vehicles’ AoI exceeds a predefined threshold and maximizing the knowledge about the network dynamics. In this regard, using a Gaussian process regression (GPR) approach, an online decentralized strategy is proposed to actively learn the network dynamics, estimate the vehicles’ future AoI, and proactively allocate resources. Simulation results show a significant improvement in terms of AoI violation probability, compared to several baselines, with a reduction of at least 50%.
Tasks	Active Learning
Published	2019-11-27
URL	https://arxiv.org/abs/1912.03359v1
PDF	https://arxiv.org/pdf/1912.03359v1.pdf
PWC	https://paperswithcode.com/paper/ultra-reliable-and-low-latency-vehicular
Repo
Framework

Utterance-level Permutation Invariant Training with Latency-controlled BLSTM for Single-channel Multi-talker Speech Separation


Title	Utterance-level Permutation Invariant Training with Latency-controlled BLSTM for Single-channel Multi-talker Speech Separation
Authors	Lu Huang, Gaofeng Cheng, Pengyuan Zhang, Yi Yang, Shumin Xu, Jiasong Sun
Abstract	Utterance-level permutation invariant training (uPIT) has achieved promising progress on single-channel multi-talker speech separation task. Long short-term memory (LSTM) and bidirectional LSTM (BLSTM) are widely used as the separation networks of uPIT, i.e. uPIT-LSTM and uPIT-BLSTM. uPIT-LSTM has lower latency but worse performance, while uPIT-BLSTM has better performance but higher latency. In this paper, we propose using latency-controlled BLSTM (LC-BLSTM) during inference to fulfill low-latency and good-performance speech separation. To find a better training strategy for BLSTM-based separation network, chunk-level PIT (cPIT) and uPIT are compared. The experimental results show that uPIT outperforms cPIT when LC-BLSTM is used during inference. It is also found that the inter-chunk speaker tracing (ST) can further improve the separation performance of uPIT-LC-BLSTM. Evaluated on the WSJ0 two-talker mixed-speech separation task, the absolute gap of signal-to-distortion ratio (SDR) between uPIT-BLSTM and uPIT-LC-BLSTM is reduced to within 0.7 dB.
Tasks	Speech Separation
Published	2019-12-25
URL	https://arxiv.org/abs/1912.11613v1
PDF	https://arxiv.org/pdf/1912.11613v1.pdf
PWC	https://paperswithcode.com/paper/utterance-level-permutation-invariant
Repo
Framework

BPMR: Bayesian Probabilistic Multivariate Ranking


Title	BPMR: Bayesian Probabilistic Multivariate Ranking
Authors	Nan Wang, Hongning Wang
Abstract	Multi-aspect user preferences are attracting wider attention in recommender systems, as they enable more detailed understanding of users’ evaluations of items. Previous studies show that incorporating multi-aspect preferences can greatly improve the performance and explainability of recommendation. However, as recommendation is essentially a ranking problem, there is no principled solution for ranking multiple aspects collectively to enhance the recommendation. In this work, we derive a multi-aspect ranking criterion. To maintain the dependency among different aspects, we propose to use a vectorized representation of multi-aspect ratings and develop a probabilistic multivariate tensor factorization framework (PMTF). The framework naturally leads to a probabilistic multi-aspect ranking criterion, which generalizes the single-aspect ranking to a multivariate fashion. Experiment results on a large multi-aspect review rating dataset confirmed the effectiveness of our solution.
Tasks	Recommendation Systems
Published	2019-09-18
URL	https://arxiv.org/abs/1909.08737v1
PDF	https://arxiv.org/pdf/1909.08737v1.pdf
PWC	https://paperswithcode.com/paper/bpmr-bayesian-probabilistic-multivariate
Repo
Framework

Colorectal Polyp Segmentation by U-Net with Dilation Convolution


Title	Colorectal Polyp Segmentation by U-Net with Dilation Convolution
Authors	Xinzi Sun, Pengfei Zhang, Dechun Wang, Yu Cao, Benyuan Liu
Abstract	Colorectal cancer (CRC) is one of the most commonly diagnosed cancers and a leading cause of cancer deaths in the United States. Colorectal polyps that grow on the intima of the colon or rectum is an important precursor for CRC. Currently, the most common way for colorectal polyp detection and precancerous pathology is the colonoscopy. Therefore, accurate colorectal polyp segmentation during the colonoscopy procedure has great clinical significance in CRC early detection and prevention. In this paper, we propose a novel end-to-end deep learning framework for the colorectal polyp segmentation. The model we design consists of an encoder to extract multi-scale semantic features and a decoder to expand the feature maps to a polyp segmentation map. We improve the feature representation ability of the encoder by introducing the dilated convolution to learn high-level semantic features without resolution reduction. We further design a simplified decoder which combines multi-scale semantic features with fewer parameters than the traditional architecture. Furthermore, we apply three post processing techniques on the output segmentation map to improve colorectal polyp detection performance. Our method achieves state-of-the-art results on CVC-ClinicDB and ETIS-Larib Polyp DB.
Tasks
Published	2019-12-26
URL	https://arxiv.org/abs/1912.11947v1
PDF	https://arxiv.org/pdf/1912.11947v1.pdf
PWC	https://paperswithcode.com/paper/colorectal-polyp-segmentation-by-u-net-with
Repo
Framework

Automatic Pulmonary Nodule Detection in CT Scans Using Convolutional Neural Networks Based on Maximum Intensity Projection


Title	Automatic Pulmonary Nodule Detection in CT Scans Using Convolutional Neural Networks Based on Maximum Intensity Projection
Authors	Sunyi Zheng, Jiapan Guo, Xiaonan Cui, Raymond N. J. Veldhuis, Matthijs Oudkerk, Peter M. A. van Ooijen
Abstract	Accurate pulmonary nodule detection is a crucial step in lung cancer screening. Computer-aided detection (CAD) systems are not routinely used by radiologists for pulmonary nodule detection in clinical practice despite their potential benefits. Maximum intensity projection (MIP) images improve the detection of pulmonary nodules in radiological evaluation with computed tomography (CT) scans. Inspired by the clinical methodology of radiologists, we aim to explore the feasibility of applying MIP images to improve the effectiveness of automatic lung nodule detection using convolutional neural networks (CNNs). We propose a CNN-based approach that takes MIP images of different slab thicknesses (5 mm, 10 mm, 15 mm) and 1 mm axial section slices as input. Such an approach augments the two-dimensional (2-D) CT slice images with more representative spatial information that helps discriminate nodules from vessels through their morphologies. Our proposed method achieves sensitivity of 92.67% with 1 false positive per scan and sensitivity of 94.19% with 2 false positives per scan for lung nodule detection on 888 scans in the LIDC-IDRI dataset. The use of thick MIP images helps the detection of small pulmonary nodules (3 mm-10 mm) and results in fewer false positives. Experimental results show that utilizing MIP images can increase the sensitivity and lower the number of false positives, which demonstrates the effectiveness and significance of the proposed MIP-based CNNs framework for automatic pulmonary nodule detection in CT scans. The proposed method also shows the potential that CNNs could gain benefits for nodule detection by combining the clinical procedure.
Tasks	Computed Tomography (CT), Lung Nodule Detection
Published	2019-04-11
URL	https://arxiv.org/abs/1904.05956v2
PDF	https://arxiv.org/pdf/1904.05956v2.pdf
PWC	https://paperswithcode.com/paper/automatic-pulmonary-nodule-detection-in-ct
Repo
Framework

Alliances and Conflict, or Conflict and Alliances? Appraising the Causal Effect of Alliances on Conflict


Title	Alliances and Conflict, or Conflict and Alliances? Appraising the Causal Effect of Alliances on Conflict
Authors	Benjamin Campbell
Abstract	The deterrent effect of military alliances is well documented and widely accepted. However, such work has typically assumed that alliances are exogenous. This is problematic as alliances may simultaneously influence the probability of conflict and be influenced by the probability of conflict. Failing to account for such endogeneity produces overly simplistic theories of alliance politics and barriers to identifying the causal effect of alliances on conflict. In this manuscript, I propose a solution to this theoretical and empirical modeling challenge. Synthesizing theories of alliance formation and the alliance-conflict relationship, I innovate an endogenous theory of alliances and conflict. I then test this theory using innovative generalized joint regression models that allow me to endogenize alliance formation on the causal path to conflict. Once doing so, I ultimately find that alliances neither deter nor provoke aggression. This has significant implications for our understanding of interstate conflict and alliance politics.
Tasks
Published	2019-08-19
URL	https://arxiv.org/abs/1908.07100v1
PDF	https://arxiv.org/pdf/1908.07100v1.pdf
PWC	https://paperswithcode.com/paper/alliances-and-conflict-or-conflict-and
Repo
Framework

Tree bark re-identification using a deep-learning feature descriptor


Title	Tree bark re-identification using a deep-learning feature descriptor
Authors	Martin Robert, Patrick Dallaire, Philippe Giguère
Abstract	The ability to visually re-identify objects is a fundamental capability in vision systems. Oftentimes, it relies on collections of visual signatures based on descriptors, such as SIFT or SURF. However, these traditional descriptors were designed for a certain domain of surface appearances and geometries (limited relief). Consequently, highly-textured surfaces such as tree bark pose a challenge to them. In turn, this makes it more difficult to use trees as identifiable landmarks for navigational purposes (robotics) or to track felled lumber along a supply chain (logistics). We thus propose to use data-driven descriptors trained on bark images for tree surface re-identification. To this effect, we collected a large dataset containing 2,400 bark images with strong illumination changes, annotated by surface and with the ability to pixel-align them. We used this dataset to sample from more than 2 million 64x64 pixel patches to train our novel local descriptors DeepBark and SqueezeBark. Our DeepBark method has shown a clear advantage against the hand-crafted descriptors SIFT and SURF. For instance, we demonstrated that DeepBark can reach a mAP of 87.2% when retrieving 11 relevant bark images, i.e. corresponding to the same physical surface, to a bark query against 7,900 images. Our work thus suggests that re-identifying tree surfaces in a challenging illuminations context is possible. We also make public our dataset, which can be used to benchmark surface re-identification techniques.
Tasks
Published	2019-12-06
URL	https://arxiv.org/abs/1912.03221v2
PDF	https://arxiv.org/pdf/1912.03221v2.pdf
PWC	https://paperswithcode.com/paper/tree-bark-re-identification-using-a-deep
Repo
Framework

Genetic Neural Architecture Search for automatic assessment of human sperm images


Title	Genetic Neural Architecture Search for automatic assessment of human sperm images
Authors	Erfan Miahi, Seyed Abolghasem Mirroshandel, Alexis Nasr
Abstract	Male infertility is a disease which affects approximately 7% of men. Sperm morphology analysis (SMA) is one of the main diagnosis methods for this problem. Manual SMA is an inexact, subjective, non-reproducible, and hard to teach process. As a result, in this paper, we introduce a novel automatic SMA based on a neural architecture search algorithm termed Genetic Neural Architecture Search (GeNAS). For this purpose, we used a collection of images called MHSMA dataset contains 1,540 sperm images which have been collected from 235 patients with infertility problems. GeNAS is a genetic algorithm that acts as a meta-controller which explores the constrained search space of plain convolutional neural network architectures. Every individual of the genetic algorithm is a convolutional neural network trained to predict morphological deformities in different segments of human sperm (head, vacuole, and acrosome), and its fitness is calculated by a novel proposed method named GeNAS-WF especially designed for noisy, low resolution, and imbalanced datasets. Also, a hashing method is used to save each trained neural architecture fitness, so we could reuse them during fitness evaluation and speed up the algorithm. Besides, in terms of running time and computation power, our proposed architecture search method is far more efficient than most of the other existing neural architecture search algorithms. Additionally, other proposed methods have been evaluated on balanced datasets, whereas GeNAS is built specifically for noisy, low quality, and imbalanced datasets which are common in the field of medical imaging. In our experiments, the best neural architecture found by GeNAS has reached an accuracy of 92.66%, 77.33%, and 77.66% in the vacuole, head, and acrosome abnormality detection, respectively. In comparison to other proposed algorithms for MHSMA dataset, GeNAS achieved state-of-the-art results.
Tasks	Anomaly Detection, Neural Architecture Search
Published	2019-09-20
URL	https://arxiv.org/abs/1909.09432v1
PDF	https://arxiv.org/pdf/1909.09432v1.pdf
PWC	https://paperswithcode.com/paper/genetic-neural-architecture-search-for
Repo
Framework

Cognitive Agent Based Simulation Model For Improving Disaster Response Procedures


Title	Cognitive Agent Based Simulation Model For Improving Disaster Response Procedures
Authors	Rohit K. Dubey, Samuel S. Sohn, Christoph Hoelscher, Mubbasir Kapadia
Abstract	In the event of a disaster, saving human lives is of utmost importance. For developing proper evacuation procedures and guidance systems, behavioural data on how people respond during panic and stress is crucial. In the absence of real human data on building evacuation, there is a need for a crowd simulator to model egress and decision-making under uncertainty. In this paper, we propose an agent-based simulation tool, which is grounded in human cognition and decision-making, for evaluating and improving the effectiveness of building evacuation procedures and guidance systems during a disaster. Specifically, we propose a predictive agent-wayfinding framework based on information theory that is applied at intersections with variable route choices where it fuses N dynamic information sources. The proposed framework can be used to visualize trajectories and prediction results (i.e., total evacuation time, number of people evacuated) for different combinations of reinforcing or contradicting information sources (i.e., signage, crowd flow, familiarity, and spatial layout). This tool can enable designers to recreate various disaster scenarios and generate simulation data for improving the evacuation procedures and existing guidance systems.
Tasks	Decision Making, Decision Making Under Uncertainty
Published	2019-10-02
URL	https://arxiv.org/abs/1910.00767v1
PDF	https://arxiv.org/pdf/1910.00767v1.pdf
PWC	https://paperswithcode.com/paper/cognitive-agent-based-simulation-model-for
Repo
Framework

Interpretable Classification from Skin Cancer Histology Slides Using Deep Learning: A Retrospective Multicenter Study


Title	Interpretable Classification from Skin Cancer Histology Slides Using Deep Learning: A Retrospective Multicenter Study
Authors	Peizhen Xie, Ke Zuo, Yu Zhang, Fangfang Li, Mingzhu Yin, Kai Lu
Abstract	For diagnosing melanoma, hematoxylin and eosin (H&E) stained tissue slides remains the gold standard. These images contain quantitative information in different magnifications. In the present study, we investigated whether deep convolutional neural networks can extract structural features of complex tissues directly from these massive size images in a patched way. In order to face the challenge arise from morphological diversity in histopathological slides, we built a multicenter database of 2241 digital whole-slide images from 1321 patients from 2008 to 2018. We trained both ResNet50 and Vgg19 using over 9.95 million patches by transferring learning, and test performance with two kinds of critical classifications: malignant melanomas versus benign nevi in separate and mixed magnification; and distinguish among nevi in maximum magnification. The CNNs achieves superior performance across both tasks, demonstrating an AI capable of classifying skin cancer in the analysis from histopathological images. For making the classifications reasonable, the visualization of CNN representations is furthermore used to identify cells between melanoma and nevi. Regions of interest (ROI) are also located which are significantly helpful, giving pathologists more support of correctly diagnosis.
Tasks
Published	2019-04-12
URL	http://arxiv.org/abs/1904.06156v1
PDF	http://arxiv.org/pdf/1904.06156v1.pdf
PWC	https://paperswithcode.com/paper/interpretable-classification-from-skin-cancer
Repo
Framework

Video from Stills: Lensless Imaging with Rolling Shutter


Title	Video from Stills: Lensless Imaging with Rolling Shutter
Authors	Nick Antipa, Patrick Oare, Emrah Bostan, Ren Ng, Laura Waller
Abstract	Because image sensor chips have a finite bandwidth with which to read out pixels, recording video typically requires a trade-off between frame rate and pixel count. Compressed sensing techniques can circumvent this trade-off by assuming that the image is compressible. Here, we propose using multiplexing optics to spatially compress the scene, enabling information about the whole scene to be sampled from a row of sensor pixels, which can be read off quickly via a rolling shutter CMOS sensor. Conveniently, such multiplexing can be achieved with a simple lensless, diffuser-based imaging system. Using sparse recovery methods, we are able to recover 140 video frames at over 4,500 frames per second, all from a single captured image with a rolling shutter sensor. Our proof-of-concept system uses easily-fabricated diffusers paired with an off-the-shelf sensor. The resulting prototype enables compressive encoding of high frame rate video into a single rolling shutter exposure, and exceeds the sampling-limited performance of an equivalent global shutter system for sufficiently sparse objects.
Tasks
Published	2019-05-30
URL	https://arxiv.org/abs/1905.13221v1
PDF	https://arxiv.org/pdf/1905.13221v1.pdf
PWC	https://paperswithcode.com/paper/video-from-stills-lensless-imaging-with
Repo
Framework