April 2, 2020

3154 words 15 mins read

Paper Group ANR 286

Paper Group ANR 286

DuDoNet++: Encoding mask projection to reduce CT metal artifacts. Saliency Enhancement using Gradient Domain Edges Merging. DNN-based Localization from Channel Estimates: Feature Design and Experimental Results. Domain Embedded Multi-model Generative Adversarial Networks for Image-based Face Inpainting. Learning to simulate and design for structura …

DuDoNet++: Encoding mask projection to reduce CT metal artifacts

Title DuDoNet++: Encoding mask projection to reduce CT metal artifacts
Authors Yuanyuan Lyu, Wei-An Lin, Jingjing Lu, S. Kevin Zhou
Abstract CT metal artifact reduction (MAR) is a notoriously challenging task because the artifacts are structured and non-local in the image domain. However, they are inherently local in the sinogram domain. DuDoNet is the state-of-the-art MAR algorithm which exploits the latter characteristic by learning to reduce artifacts in the sinogram and image domain jointly. By design, DuDoNet treats the metal-affected regions in sinogram as missing and replaces them with the surrogate data generated by a neural network. Since fine-grained details within the metal-affected regions are completely ignored, the artifact-reduced CT images by DuDoNet tend to be over-smoothed and distorted. In this work, we investigate the issue by theoretical derivation. We propose to address the problem by (1) retaining the metal-affected regions in sinogram and (2) replacing the binarized metal trace with the metal mask projection such that the geometry information of metal implants is encoded. Extensive experiments on simulated datasets and expert evaluations on clinical images demonstrate that our network called DuDoNet++ yields anatomically more precise artifact-reduced images than DuDoNet, especially when the metallic objects are large.
Tasks Computed Tomography (CT), Metal Artifact Reduction
Published 2020-01-02
URL https://arxiv.org/abs/2001.00340v2
PDF https://arxiv.org/pdf/2001.00340v2.pdf
PWC https://paperswithcode.com/paper/dudonet-encoding-mask-projection-to-reduce-ct

Saliency Enhancement using Gradient Domain Edges Merging

Title Saliency Enhancement using Gradient Domain Edges Merging
Authors Dominique Beaini, Sofiane Achiche, Alexandre Duperre, Maxime Raison
Abstract In recent years, there has been a rapid progress in solving the binary problems in computer vision, such as edge detection which finds the boundaries of an image and salient object detection which finds the important object in an image. This progress happened thanks to the rise of deep-learning and convolutional neural networks (CNN) which allow to extract complex and abstract features. However, edge detection and saliency are still two different fields and do not interact together, although it is intuitive for a human to detect salient objects based on its boundaries. Those features are not well merged in a CNN because edges and surfaces do not intersect since one feature represents a region while the other represents boundaries between different regions. In the current work, the main objective is to develop a method to merge the edges with the saliency maps to improve the performance of the saliency. Hence, we developed the gradient-domain merging (GDM) which can be used to quickly combine the image-domain information of salient object detection with the gradient-domain information of the edge detection. This leads to our proposed saliency enhancement using edges (SEE) with an average improvement of the F-measure of at least 3.4 times higher on the DUT-OMRON dataset and 6.6 times higher on the ECSSD dataset, when compared to competing algorithm such as denseCRF and BGOF. The SEE algorithm is split into 2 parts, SEE-Pre for preprocessing and SEE-Post pour postprocessing.
Tasks Edge Detection, Object Detection, Salient Object Detection
Published 2020-02-11
URL https://arxiv.org/abs/2002.04380v1
PDF https://arxiv.org/pdf/2002.04380v1.pdf
PWC https://paperswithcode.com/paper/saliency-enhancement-using-gradient-domain

DNN-based Localization from Channel Estimates: Feature Design and Experimental Results

Title DNN-based Localization from Channel Estimates: Feature Design and Experimental Results
Authors Paul Ferrand, Alexis Decurninge, Maxime Guillaud
Abstract We consider the use of deep neural networks (DNNs) in the context of channel state information (CSI)-based localization for Massive MIMO cellular systems. We discuss the practical impairments that are likely to be present in practical CSI estimates, and introduce a principled approach to feature design for CSI-based DNN applications based on the objective of making the features invariant to the considered impairments. We demonstrate the efficiency of this approach by applying it to a dataset constituted of geo-tagged CSI measured in an outdoors campus environment, and training a DNN to estimate the position of the UE on the basis of the CSI. We provide an experimental evaluation of several aspects of that learning approach, including localization accuracy, generalization capability, and data aging.
Published 2020-03-20
URL https://arxiv.org/abs/2004.00363v1
PDF https://arxiv.org/pdf/2004.00363v1.pdf
PWC https://paperswithcode.com/paper/dnn-based-localization-from-channel-estimates

Domain Embedded Multi-model Generative Adversarial Networks for Image-based Face Inpainting

Title Domain Embedded Multi-model Generative Adversarial Networks for Image-based Face Inpainting
Authors Xian Zhang, Xin Wang, Bin Kong, Youbing Yin, Qi Song, Siwei Lyu, Jiancheng Lv, Canghong Shi, Xiaojie Li
Abstract Prior knowledge of face shape and location plays an important role in face inpainting. However, traditional facing inpainting methods mainly focus on the generated image resolution of the missing portion but without consideration of the special particularities of the human face explicitly and generally produce discordant facial parts. To solve this problem, we present a stable variational latent generative model for large inpainting of face images. We firstly represent only face regions with the latent variable space but simultaneously constraint the random vectors to offer control over the distribution of latent variables, and combine with the non-face parts textures to generate a face image with plausible contents. Two adversarial discriminators are finally used to judge whether the generated distribution is close to the real distribution or not. It can not only synthesize novel image structures but also explicitly utilize the latent space with Eigenfaces to make better predictions. Furthermore, our work better evaluates the side face impainting problem. Experiments on both CelebA and CelebA-HQ face datasets demonstrate that our proposed approach generates higher quality inpainting results than existing ones.
Tasks Facial Inpainting
Published 2020-02-05
URL https://arxiv.org/abs/2002.02909v1
PDF https://arxiv.org/pdf/2002.02909v1.pdf
PWC https://paperswithcode.com/paper/domain-embedded-multi-model-generative

Learning to simulate and design for structural engineering

Title Learning to simulate and design for structural engineering
Authors Kai-Hung Chang, Chin-Yi Cheng
Abstract In the architecture and construction industries, structural design for large buildings has always been laborious, time-consuming, and difficult to optimize. It is an iterative process that involves two steps: analyzing the current structural design by a slow and computationally expensive simulation, and then manually revising the design based on professional experience and rules. In this work, we propose an end-to-end learning pipeline to solve the size design optimization problem, which is to design the optimal cross-sections for columns and beams, given the design objectives and building code as constraints. We pre-train a graph neural network as a surrogate model to not only replace the structural simulation for speed but also use its differentiable nature to provide gradient signals to the other graph neural network for size optimization. Our results show that the pre-trained surrogate model can predict simulation results accurately, and the trained optimization model demonstrates the capability of designing convincing cross-section designs for buildings under various scenarios.
Published 2020-03-20
URL https://arxiv.org/abs/2003.09103v1
PDF https://arxiv.org/pdf/2003.09103v1.pdf
PWC https://paperswithcode.com/paper/learning-to-simulate-and-design-for

Robust Market Making via Adversarial Reinforcement Learning

Title Robust Market Making via Adversarial Reinforcement Learning
Authors Thomas Spooner, Rahul Savani
Abstract We show that adversarial reinforcement learning (ARL) can be used to produce market marking agents that are robust to adversarial and adaptively chosen market conditions. To apply ARL, we turn the well-studied single-agent model of Avellaneda and Stoikov [2008] into a discrete-time zero-sum game between a market maker and adversary, a proxy for other market participants who would like to profit at the market maker’s expense. We empirically compare two conventional single-agent RL agents with ARL, and show that our ARL approach leads to: 1) the emergence of naturally risk-averse behaviour without constraints or domain-specific penalties; 2) significant improvements in performance across a set of standard metrics, evaluated with or without an adversary in the test environment, and; 3) improved robustness to model uncertainty. We empirically demonstrate that our ARL method consistently converges, and we prove for several special cases that the profiles that we converge to are Nash equilibria in a corresponding simplified single-stage game.
Published 2020-03-03
URL https://arxiv.org/abs/2003.01820v1
PDF https://arxiv.org/pdf/2003.01820v1.pdf
PWC https://paperswithcode.com/paper/robust-market-making-via-adversarial

Deep-URL: A Model-Aware Approach To Blind Deconvolution Based On Deep Unfolded Richardson-Lucy Network

Title Deep-URL: A Model-Aware Approach To Blind Deconvolution Based On Deep Unfolded Richardson-Lucy Network
Authors Chirag Agarwal, Shahin Khobahi, Arindam Bose, Mojtaba Soltanalian, Dan Schonfeld
Abstract The lack of interpretability in current deep learning models causes serious concerns as they are extensively used for various life-critical applications. Hence, it is of paramount importance to develop interpretable deep learning models. In this paper, we consider the problem of blind deconvolution and propose a novel model-aware deep architecture that allows for the recovery of both the blur kernel and the sharp image from the blurred image. In particular, we propose the Deep Unfolded Richardson-Lucy (Deep-URL) framework – an interpretable deep-learning architecture that can be seen as an amalgamation of classical estimation technique and deep neural network, and consequently leads to improved performance. Our numerical investigations demonstrate significant improvement compared to state-of-the-art algorithms.
Published 2020-02-03
URL https://arxiv.org/abs/2002.01053v2
PDF https://arxiv.org/pdf/2002.01053v2.pdf
PWC https://paperswithcode.com/paper/deep-url-a-model-aware-approach-to-blind

Estimating Multiple Precision Matrices with Cluster Fusion Regularization

Title Estimating Multiple Precision Matrices with Cluster Fusion Regularization
Authors Bradley S. Price, Aaron J. Molstad, Ben Sherwood
Abstract We propose a penalized likelihood framework for estimating multiple precision matrices from different classes. Most existing methods either incorporate no information on relationships between the precision matrices, or require this information be known a priori. The framework proposed in this article allows for simultaneous estimation of the precision matrices and relationships between the precision matrices, jointly. Sparse and non-sparse estimators are proposed, both of which require solving a non-convex optimization problem. To compute our proposed estimators, we use an iterative algorithm which alternates between a convex optimization problem solved by blockwise coordinate descent and a k-means clustering problem. Blockwise updates for computing the sparse estimator require solving an elastic net penalized precision matrix estimation problem, which we solve using a proximal gradient descent algorithm. We prove that this subalgorithm has a linear rate of convergence. In simulation studies and two real data applications, we show that our method can outperform competitors that ignore relevant relationships between precision matrices and performs similarly to methods which use prior information often uknown in practice.
Published 2020-03-01
URL https://arxiv.org/abs/2003.00371v1
PDF https://arxiv.org/pdf/2003.00371v1.pdf
PWC https://paperswithcode.com/paper/estimating-multiple-precision-matrices-with

Network-based models for social recommender systems

Title Network-based models for social recommender systems
Authors Antonia Godoy-Lorite, Roger Guimera, Marta Sales-Pardo
Abstract With the overwhelming online products available in recent years, there is an increasing need to filter and deliver relevant personalized advice for users. Recommender systems solve this problem by modeling and predicting individual preferences for a great variety of items such as movies, books or research articles. In this chapter, we explore rigorous network-based models that outperform leading approaches for recommendation. The network models we consider are based on the explicit assumption that there are groups of individuals and of items, and that the preferences of an individual for an item are determined only by their group memberships. The accurate prediction of individual user preferences over items can be accomplished by different methodologies, such as Monte Carlo sampling or Expectation-Maximization methods, the latter resulting in a scalable algorithm which is suitable for large datasets.
Tasks Recommendation Systems
Published 2020-02-10
URL https://arxiv.org/abs/2002.03700v1
PDF https://arxiv.org/pdf/2002.03700v1.pdf
PWC https://paperswithcode.com/paper/network-based-models-for-social-recommender

Correcting for Selection Bias in Learning-to-rank Systems

Title Correcting for Selection Bias in Learning-to-rank Systems
Authors Zohreh Ovaisi, Ragib Ahsan, Yifan Zhang, Kathryn Vasilaky, Elena Zheleva
Abstract Click data collected by modern recommendation systems are an important source of observational data that can be utilized to train learning-to-rank (LTR) systems. However, these data suffer from a number of biases that can result in poor performance for LTR systems. Recent methods for bias correction in such systems mostly focus on position bias, the fact that higher ranked results (e.g., top search engine results) are more likely to be clicked even if they are not the most relevant results given a user’s query. Less attention has been paid to correcting for selection bias, which occurs because clicked documents are reflective of what documents have been shown to the user in the first place. Here, we propose new counterfactual approaches which adapt Heckman’s two-stage method and accounts for selection and position bias in LTR systems. Our empirical evaluation shows that our proposed methods are much more robust to noise and have better accuracy compared to existing unbiased LTR algorithms, especially when there is moderate to no position bias.
Tasks Learning-To-Rank, Recommendation Systems
Published 2020-01-29
URL https://arxiv.org/abs/2001.11358v1
PDF https://arxiv.org/pdf/2001.11358v1.pdf
PWC https://paperswithcode.com/paper/correcting-for-selection-bias-in-learning-to

Estimating Error and Bias in Offline Evaluation Results

Title Estimating Error and Bias in Offline Evaluation Results
Authors Mucun Tian, Michael D. Ekstrand
Abstract Offline evaluations of recommender systems attempt to estimate users’ satisfaction with recommendations using static data from prior user interactions. These evaluations provide researchers and developers with first approximations of the likely performance of a new system and help weed out bad ideas before presenting them to users. However, offline evaluation cannot accurately assess novel, relevant recommendations, because the most novel items were previously unknown to the user, so they are missing from the historical data and cannot be judged as relevant. We present a simulation study to estimate the error that such missing data causes in commonly-used evaluation metrics in order to assess its prevalence and impact. We find that missing data in the rating or observation process causes the evaluation protocol to systematically mis-estimate metric values, and in some cases erroneously determine that a popularity-based recommender outperforms even a perfect personalized recommender. Substantial breakthroughs in recommendation quality, therefore, will be difficult to assess with existing offline techniques.
Tasks Recommendation Systems
Published 2020-01-26
URL https://arxiv.org/abs/2001.09455v1
PDF https://arxiv.org/pdf/2001.09455v1.pdf
PWC https://paperswithcode.com/paper/estimating-error-and-bias-in-offline

Bending Loss Regularized Network for Nuclei Segmentation in Histopathology Images

Title Bending Loss Regularized Network for Nuclei Segmentation in Histopathology Images
Authors Haotian Wang, Min Xian, Aleksandar Vakanski
Abstract Separating overlapped nuclei is a major challenge in histopathology image analysis. Recently published approaches have achieved promising overall performance on public datasets; however, their performance in segmenting overlapped nuclei are limited. To address the issue, we propose the bending loss regularized network for nuclei segmentation. The proposed bending loss defines high penalties to contour points with large curvatures, and applies small penalties to contour points with small curvature. Minimizing the bending loss can avoid generating contours that encompass multiple nuclei. The proposed approach is validated on the MoNuSeg dataset using five quantitative metrics. It outperforms six state-of-the-art approaches on the following metrics: Aggregate Jaccard Index, Dice, Recognition Quality, and Pan-optic Quality.
Published 2020-02-03
URL https://arxiv.org/abs/2002.01020v1
PDF https://arxiv.org/pdf/2002.01020v1.pdf
PWC https://paperswithcode.com/paper/bending-loss-regularized-network-for-nuclei

Action Localization through Continual Predictive Learning

Title Action Localization through Continual Predictive Learning
Authors Sathyanarayanan N. Aakur, Sudeep Sarkar
Abstract The problem of action recognition involves locating the action in the video, both over time and spatially in the image. The dominant current approaches use supervised learning to solve this problem, and require large amounts of annotated training data, in the form of frame-level bounding box annotations around the region of interest. In this paper, we present a new approach based on continual learning that uses feature-level predictions for self-supervision. It does not require any training annotations in terms of frame-level bounding boxes. The approach is inspired by cognitive models of visual event perception that propose a prediction-based approach to event understanding. We use a stack of LSTMs coupled with CNN encoder, along with novel attention mechanisms, to model the events in the video and use this model to predict high-level features for the future frames. The prediction errors are used to continuously learn the parameters of the models. This self-supervised framework is not complicated as other approaches but is very effective in learning robust visual representations for both labeling and localization. It should be noted that the approach outputs in a streaming fashion, requiring only a single pass through the video, making it amenable for real-time processing. We demonstrate this on three datasets - UCF Sports, JHMDB, and THUMOS’13 and show that the proposed approach outperforms weakly-supervised and unsupervised baselines and obtains competitive performance compared to fully supervised baselines. Finally, we show that the proposed framework can generalize to egocentric videos and obtain state-of-the-art results in unsupervised gaze prediction.
Tasks Action Localization, Continual Learning, Gaze Prediction
Published 2020-03-26
URL https://arxiv.org/abs/2003.12185v1
PDF https://arxiv.org/pdf/2003.12185v1.pdf
PWC https://paperswithcode.com/paper/action-localization-through-continual

Optimization-Based MCMC Methods for Nonlinear Hierarchical Statistical Inverse Problems

Title Optimization-Based MCMC Methods for Nonlinear Hierarchical Statistical Inverse Problems
Authors Johnathan Bardsley, Tiangang Cui
Abstract In many hierarchical inverse problems, not only do we want to estimate high- or infinite-dimensional model parameters in the parameter-to-observable maps, but we also have to estimate hyperparameters that represent critical assumptions in the statistical and mathematical modeling processes. As a joint effect of high-dimensionality, nonlinear dependence, and non-concave structures in the joint posterior posterior distribution over model parameters and hyperparameters, solving inverse problems in the hierarchical Bayesian setting poses a significant computational challenge. In this work, we aim to develop scalable optimization-based Markov chain Monte Carlo (MCMC) methods for solving hierarchical Bayesian inverse problems with nonlinear parameter-to-observable maps and a broader class of hyperparameters. Our algorithmic development is based on the recently developed scalable randomize-then-optimize (RTO) method [4] for exploring the high- or infinite-dimensional model parameter space. By using RTO either as a proposal distribution in a Metropolis-within-Gibbs update or as a biasing distribution in the pseudo-marginal MCMC [2], we are able to design efficient sampling tools for hierarchical Bayesian inversion. In particular, the integration of RTO and the pseudo-marginal MCMC has sampling performance robust to model parameter dimensions. We also extend our methods to nonlinear inverse problems with Poisson-distributed measurements. Numerical examples in PDE-constrained inverse problems and positron emission tomography (PET) are used to demonstrate the performance of our methods.
Published 2020-02-15
URL https://arxiv.org/abs/2002.06358v1
PDF https://arxiv.org/pdf/2002.06358v1.pdf
PWC https://paperswithcode.com/paper/optimization-based-mcmc-methods-for-nonlinear

How do Data Science Workers Collaborate? Roles, Workflows, and Tools

Title How do Data Science Workers Collaborate? Roles, Workflows, and Tools
Authors Amy X. Zhang, Michael Muller, Dakuo Wang
Abstract Today, the prominence of data science within organizations has given rise to teams of data science workers collaborating on extracting insights from data, as opposed to individual data scientists working alone. However, we still lack a deep understanding of how data science workers collaborate in practice. In this work, we conducted an online survey with 183 participants who work in various aspects of data science. We focused on their reported interactions with each other (e.g., managers with engineers) and with different tools (e.g., Jupyter Notebook). We found that data science teams are extremely collaborative and work with a variety of stakeholders and tools during the six common steps of a data science workflow (e.g., clean data and train model). We also found that the collaborative practices workers employ, such as documentation, vary according to the kinds of tools they use. Based on these findings, we discuss design implications for supporting data science team collaborations and future research directions.
Published 2020-01-18
URL https://arxiv.org/abs/2001.06684v2
PDF https://arxiv.org/pdf/2001.06684v2.pdf
PWC https://paperswithcode.com/paper/how-do-data-science-workers-collaborate-roles
comments powered by Disqus