July 27, 2019

3159 words 15 mins read

Paper Group ANR 492

Deep Structure for end-to-end inverse rendering. Causal Generative Neural Networks. Classification of MRI data using Deep Learning and Gaussian Process-based Model Selection. Fashioning with Networks: Neural Style Transfer to Design Clothes. Bi-Level Online Control without Regret. Texture Object Segmentation Based on Affine Invariant Texture Detect …

Deep Structure for end-to-end inverse rendering


Title	Deep Structure for end-to-end inverse rendering
Authors	Shima Kamyab, Ali Ghodsi, S. Zohreh Azimifar
Abstract	Inverse rendering in a 3D format denoted to recovering the 3D properties of a scene given 2D input image(s) and is typically done using 3D Morphable Model (3DMM) based methods from single view images. These models formulate each face as a weighted combination of some basis vectors extracted from the training data. In this paper a deep framework is proposed in which the coefficients and basis vectors are computed by training an autoencoder network and a Convolutional Neural Network (CNN) simultaneously. The idea is to find a common cause which can be mapped to both the 3D structure and corresponding 2D image using deep networks. The empirical results verify the power of deep framework in finding accurate 3D shapes of human faces from their corresponding 2D images on synthetic datasets of human faces.
Tasks
Published	2017-08-25
URL	http://arxiv.org/abs/1708.08998v1
PDF	http://arxiv.org/pdf/1708.08998v1.pdf
PWC	https://paperswithcode.com/paper/deep-structure-for-end-to-end-inverse
Repo
Framework

Causal Generative Neural Networks


Title	Causal Generative Neural Networks
Authors	Olivier Goudet, Diviyan Kalainathan, Philippe Caillou, Isabelle Guyon, David Lopez-Paz, Michèle Sebag
Abstract	We present Causal Generative Neural Networks (CGNNs) to learn functional causal models from observational data. CGNNs leverage conditional independencies and distributional asymmetries to discover bivariate and multivariate causal structures. CGNNs make no assumption regarding the lack of confounders, and learn a differentiable generative model of the data by using backpropagation. Extensive experiments show their good performances comparatively to the state of the art in observational causal discovery on both simulated and real data, with respect to cause-effect inference, v-structure identification, and multivariate causal discovery.
Tasks	Causal Discovery
Published	2017-11-24
URL	http://arxiv.org/abs/1711.08936v2
PDF	http://arxiv.org/pdf/1711.08936v2.pdf
PWC	https://paperswithcode.com/paper/causal-generative-neural-networks
Repo
Framework

Classification of MRI data using Deep Learning and Gaussian Process-based Model Selection


Title	Classification of MRI data using Deep Learning and Gaussian Process-based Model Selection
Authors	Hadrien Bertrand, Matthieu Perrot, Roberto Ardon, Isabelle Bloch
Abstract	The classification of MRI images according to the anatomical field of view is a necessary task to solve when faced with the increasing quantity of medical images. In parallel, advances in deep learning makes it a suitable tool for computer vision problems. Using a common architecture (such as AlexNet) provides quite good results, but not sufficient for clinical use. Improving the model is not an easy task, due to the large number of hyper-parameters governing both the architecture and the training of the network, and to the limited understanding of their relevance. Since an exhaustive search is not tractable, we propose to optimize the network first by random search, and then by an adaptive search based on Gaussian Processes and Probability of Improvement. Applying this method on a large and varied MRI dataset, we show a substantial improvement between the baseline network and the final one (up to 20% for the most difficult classes).
Tasks	Gaussian Processes, Model Selection
Published	2017-01-16
URL	http://arxiv.org/abs/1701.04355v1
PDF	http://arxiv.org/pdf/1701.04355v1.pdf
PWC	https://paperswithcode.com/paper/classification-of-mri-data-using-deep
Repo
Framework

Fashioning with Networks: Neural Style Transfer to Design Clothes


Title	Fashioning with Networks: Neural Style Transfer to Design Clothes
Authors	Prutha Date, Ashwinkumar Ganesan, Tim Oates
Abstract	Convolutional Neural Networks have been highly successful in performing a host of computer vision tasks such as object recognition, object detection, image segmentation and texture synthesis. In 2015, Gatys et. al [7] show how the style of a painter can be extracted from an image of the painting and applied to another normal photograph, thus recreating the photo in the style of the painter. The method has been successfully applied to a wide range of images and has since spawned multiple applications and mobile apps. In this paper, the neural style transfer algorithm is applied to fashion so as to synthesize new custom clothes. We construct an approach to personalize and generate new custom clothes based on a users preference and by learning the users fashion choices from a limited set of clothes from their closet. The approach is evaluated by analyzing the generated images of clothes and how well they align with the users fashion style.
Tasks	Object Detection, Object Recognition, Semantic Segmentation, Style Transfer, Texture Synthesis
Published	2017-07-31
URL	http://arxiv.org/abs/1707.09899v1
PDF	http://arxiv.org/pdf/1707.09899v1.pdf
PWC	https://paperswithcode.com/paper/fashioning-with-networks-neural-style
Repo
Framework

Bi-Level Online Control without Regret


Title	Bi-Level Online Control without Regret
Authors	Andrey Bernstein
Abstract	This paper considers a bi-level discrete-time control framework with real-time constraints, consisting of several local controllers and a central controller. The objective is to bridge the gap between the online convex optimization and real-time control literature by proposing an online control algorithm with small dynamic regret, which is a natural performance criterion in nonstationary environments related to real-time control problems. We illustrate how the proposed algorithm can be applied to real-time control of power setpoints in an electrical grid.
Tasks
Published	2017-02-18
URL	http://arxiv.org/abs/1702.05548v1
PDF	http://arxiv.org/pdf/1702.05548v1.pdf
PWC	https://paperswithcode.com/paper/bi-level-online-control-without-regret
Repo
Framework

Texture Object Segmentation Based on Affine Invariant Texture Detection


Title	Texture Object Segmentation Based on Affine Invariant Texture Detection
Authors	Jianwei Zhang, Xu Chen, Xuezhong Xiao
Abstract	To solve the issue of segmenting rich texture images, a novel detection methods based on the affine invariable principle is proposed. Considering the similarity between the texture areas, we first take the affine transform to get numerous shapes, and utilize the KLT algorithm to verify the similarity. The transforms include rotation, proportional transformation and perspective deformation to cope with a variety of situations. Then we propose an improved LBP method combining canny edge detection to handle the boundary in the segmentation process. Moreover, human-computer interaction of this method which helps splitting the matched texture area from the original images is user-friendly.
Tasks	Edge Detection, Semantic Segmentation
Published	2017-12-23
URL	http://arxiv.org/abs/1712.08776v1
PDF	http://arxiv.org/pdf/1712.08776v1.pdf
PWC	https://paperswithcode.com/paper/texture-object-segmentation-based-on-affine
Repo
Framework

Scale Up Event Extraction Learning via Automatic Training Data Generation


Title	Scale Up Event Extraction Learning via Automatic Training Data Generation
Authors	Ying Zeng, Yansong Feng, Rong Ma, Zheng Wang, Rui Yan, Chongde Shi, Dongyan Zhao
Abstract	The task of event extraction has long been investigated in a supervised learning paradigm, which is bound by the number and the quality of the training instances. Existing training data must be manually generated through a combination of expert domain knowledge and extensive human involvement. However, due to drastic efforts required in annotating text, the resultant datasets are usually small, which severally affects the quality of the learned model, making it hard to generalize. Our work develops an automatic approach for generating training data for event extraction. Our approach allows us to scale up event extraction training instances from thousands to hundreds of thousands, and it does this at a much lower cost than a manual approach. We achieve this by employing distant supervision to automatically create event annotations from unlabelled text using existing structured knowledge bases or tables.We then develop a neural network model with post inference to transfer the knowledge extracted from structured knowledge bases to automatically annotate typed events with corresponding arguments in text.We evaluate our approach by using the knowledge extracted from Freebase to label texts from Wikipedia articles. Experimental results show that our approach can generate a large number of high quality training instances. We show that this large volume of training data not only leads to a better event extractor, but also allows us to detect multiple typed events.
Tasks
Published	2017-12-11
URL	http://arxiv.org/abs/1712.03665v1
PDF	http://arxiv.org/pdf/1712.03665v1.pdf
PWC	https://paperswithcode.com/paper/scale-up-event-extraction-learning-via
Repo
Framework

Characterizing the structural diversity of complex networks across domains


Title	Characterizing the structural diversity of complex networks across domains
Authors	Kansuke Ikehara, Aaron Clauset
Abstract	The structure of complex networks has been of interest in many scientific and engineering disciplines over the decades. A number of studies in the field have been focused on finding the common properties among different kinds of networks such as heavy-tail degree distribution, small-worldness and modular structure and they have tried to establish a theory of structural universality in complex networks. However, there is no comprehensive study of network structure across a diverse set of domains in order to explain the structural diversity we observe in the real-world networks. In this paper, we study 986 real-world networks of diverse domains ranging from ecological food webs to online social networks along with 575 networks generated from four popular network models. Our study utilizes a number of machine learning techniques such as random forest and confusion matrix in order to show the relationships among network domains in terms of network structure. Our results indicate that there are some partitions of network categories in which networks are hard to distinguish based purely on network structure. We have found that these partitions of network categories tend to have similar underlying functions, constraints and/or generative mechanisms of networks even though networks in the same partition have different origins, e.g., biological processes, results of engineering by human being, etc. This suggests that the origin of a network, whether it’s biological, technological or social, may not necessarily be a decisive factor of the formation of similar network structure. Our findings shed light on the possible direction along which we could uncover the hidden principles for the structural diversity of complex networks.
Tasks
Published	2017-10-31
URL	http://arxiv.org/abs/1710.11304v1
PDF	http://arxiv.org/pdf/1710.11304v1.pdf
PWC	https://paperswithcode.com/paper/characterizing-the-structural-diversity-of
Repo
Framework

A survey of exemplar-based texture synthesis


Title	A survey of exemplar-based texture synthesis
Authors	Lara Raad, Axel Davy, Agnès Desolneux, Jean-Michel Morel
Abstract	Exemplar-based texture synthesis is the process of generating, from an input sample, new texture images of arbitrary size and which are perceptually equivalent to the sample. The two main approaches are statistics-based methods and patch re-arrangement methods. In the first class, a texture is characterized by a statistical signature; then, a random sampling conditioned to this signature produces genuinely different texture images. The second class boils down to a clever “copy-paste” procedure, which stitches together large regions of the sample. Hybrid methods try to combine ideas from both approaches to avoid their hurdles. The recent approaches using convolutional neural networks fit to this classification, some being statistical and others performing patch re-arrangement in the feature space. They produce impressive synthesis on various kinds of textures. Nevertheless, we found that most real textures are organized at multiple scales, with global structures revealed at coarse scales and highly varying details at finer ones. Thus, when confronted with large natural images of textures the results of state-of-the-art methods degrade rapidly, and the problem of modeling them remains wide open.
Tasks	Texture Synthesis
Published	2017-07-22
URL	http://arxiv.org/abs/1707.07184v2
PDF	http://arxiv.org/pdf/1707.07184v2.pdf
PWC	https://paperswithcode.com/paper/a-survey-of-exemplar-based-texture-synthesis
Repo
Framework

Manifold Regularization for Kernelized LSTD


Title	Manifold Regularization for Kernelized LSTD
Authors	Xinyan Yan, Krzysztof Choromanski, Byron Boots, Vikas Sindhwani
Abstract	Policy evaluation or value function or Q-function approximation is a key procedure in reinforcement learning (RL). It is a necessary component of policy iteration and can be used for variance reduction in policy gradient methods. Therefore its quality has a significant impact on most RL algorithms. Motivated by manifold regularized learning, we propose a novel kernelized policy evaluation method that takes advantage of the intrinsic geometry of the state space learned from data, in order to achieve better sample efficiency and higher accuracy in Q-function approximation. Applying the proposed method in the Least-Squares Policy Iteration (LSPI) framework, we observe superior performance compared to widely used parametric basis functions on two standard benchmarks in terms of policy quality.
Tasks	Policy Gradient Methods
Published	2017-10-15
URL	http://arxiv.org/abs/1710.05387v1
PDF	http://arxiv.org/pdf/1710.05387v1.pdf
PWC	https://paperswithcode.com/paper/manifold-regularization-for-kernelized-lstd
Repo
Framework

Cross-Language Question Re-Ranking


Title	Cross-Language Question Re-Ranking
Authors	Giovanni Da San Martino, Salvatore Romeo, Alberto Barron-Cedeno, Shafiq Joty, Lluis Marquez, Alessandro Moschitti, Preslav Nakov
Abstract	We study how to find relevant questions in community forums when the language of the new questions is different from that of the existing questions in the forum. In particular, we explore the Arabic-English language pair. We compare a kernel-based system with a feed-forward neural network in a scenario where a large parallel corpus is available for training a machine translation system, bilingual dictionaries, and cross-language word embeddings. We observe that both approaches degrade the performance of the system when working on the translated text, especially the kernel-based system, which depends heavily on a syntactic kernel. We address this issue using a cross-language tree kernel, which compares the original Arabic tree to the English trees of the related questions. We show that this kernel almost closes the performance gap with respect to the monolingual system. On the neural network side, we use the parallel corpus to train cross-language embeddings, which we then use to represent the Arabic input and the English related questions in the same space. The results also improve to close to those of the monolingual neural network. Overall, the kernel system shows a better performance compared to the neural network in all cases.
Tasks	Machine Translation, Word Embeddings
Published	2017-10-04
URL	http://arxiv.org/abs/1710.01487v1
PDF	http://arxiv.org/pdf/1710.01487v1.pdf
PWC	https://paperswithcode.com/paper/cross-language-question-re-ranking
Repo
Framework

Temporal anomaly detection: calibrating the surprise


Title	Temporal anomaly detection: calibrating the surprise
Authors	Eyal Gutflaish, Aryeh Kontorovich, Sivan Sabato, Ofer Biller, Oded Sofer
Abstract	We propose a hybrid approach to temporal anomaly detection in access data of users to databases — or more generally, any kind of subject-object co-occurrence data. We consider a high-dimensional setting that also requires fast computation at test time. Our methodology identifies anomalies based on a single stationary model, instead of requiring a full temporal one, which would be prohibitive in this setting. We learn a low-rank stationary model from the training data, and then fit a regression model for predicting the expected likelihood score of normal access patterns in the future. The disparity between the predicted likelihood score and the observed one is used to assess the `surprise’ at test time. This approach enables calibration of the anomaly score, so that time-varying normal behavior patterns are not considered anomalous. We provide a detailed description of the algorithm, including a convergence analysis, and report encouraging empirical results. One of the data sets that we tested, TDA, is new for the public domain. It consists of two months’ worth of database access records from a live system. Our code is publicly available at https://github.com/eyalgut/TLR_anomaly_detection.git. The TDA data set is available at https://www.kaggle.com/eyalgut/binary-traffic-matrices. \|
Tasks	Anomaly Detection, Calibration
Published	2017-05-29
URL	http://arxiv.org/abs/1705.10085v2
PDF	http://arxiv.org/pdf/1705.10085v2.pdf
PWC	https://paperswithcode.com/paper/temporal-anomaly-detection-calibrating-the
Repo
Framework

Part-to-whole Registration of Histology and MRI using Shape Elements


Title	Part-to-whole Registration of Histology and MRI using Shape Elements
Authors	Jonas Pichat, Juan Eugenio Iglesias, Sotiris Nousias, Tarek Yousry, Sebastien Ourselin, Marc Modat
Abstract	Image registration between histology and magnetic resonance imaging (MRI) is a challenging task due to differences in structural content and contrast. Too thick and wide specimens cannot be processed all at once and must be cut into smaller pieces. This dramatically increases the complexity of the problem, since each piece should be individually and manually pre-aligned. To the best of our knowledge, no automatic method can reliably locate such piece of tissue within its respective whole in the MRI slice, and align it without any prior information. We propose here a novel automatic approach to the joint problem of multimodal registration between histology and MRI, when only a fraction of tissue is available from histology. The approach relies on the representation of images using their level lines so as to reach contrast invariance. Shape elements obtained via the extraction of bitangents are encoded in a projective-invariant manner, which permits the identification of common pieces of curves between two images. We evaluated the approach on human brain histology and compared resulting alignments against manually annotated ground truths. Considering the complexity of the brain folding patterns, preliminary results are promising and suggest the use of characteristic and meaningful shape elements for improved robustness and efficiency.
Tasks	Image Registration
Published	2017-08-27
URL	http://arxiv.org/abs/1708.08117v1
PDF	http://arxiv.org/pdf/1708.08117v1.pdf
PWC	https://paperswithcode.com/paper/part-to-whole-registration-of-histology-and
Repo
Framework

FHEDN: A based on context modeling Feature Hierarchy Encoder-Decoder Network for face detection


Title	FHEDN: A based on context modeling Feature Hierarchy Encoder-Decoder Network for face detection
Authors	Zexun Zhou, Zhongshi He, Ziyu Chen, Yuanyuan Jia, Haiyan Wang, Jinglong Du, Dingding Chen
Abstract	Because of affected by weather conditions, camera pose and range, etc. Objects are usually small, blur, occluded and diverse pose in the images gathered from outdoor surveillance cameras or access control system. It is challenging and important to detect faces precisely for face recognition system in the field of public security. In this paper, we design a based on context modeling structure named Feature Hierarchy Encoder-Decoder Network for face detection(FHEDN), which can detect small, blur and occluded face with hierarchy by hierarchy from the end to the beginning likes encoder-decoder in a single network. The proposed network is consist of multiple context modeling and prediction modules, which are in order to detect small, blur, occluded and diverse pose faces. In addition, we analyse the influence of distribution of training set, scale of default box and receipt field size to detection performance in implement stage. Demonstrated by experiments, Our network achieves promising performance on WIDER FACE and FDDB benchmarks.
Tasks	Face Detection, Face Recognition
Published	2017-12-11
URL	http://arxiv.org/abs/1712.03687v1
PDF	http://arxiv.org/pdf/1712.03687v1.pdf
PWC	https://paperswithcode.com/paper/fhedn-a-based-on-context-modeling-feature
Repo
Framework

Evolving Ensemble Fuzzy Classifier


Title	Evolving Ensemble Fuzzy Classifier
Authors	Mahardhika Pratama, Witold Pedrycz, Edwin Lughofer
Abstract	The concept of ensemble learning offers a promising avenue in learning from data streams under complex environments because it addresses the bias and variance dilemma better than its single model counterpart and features a reconfigurable structure, which is well suited to the given context. While various extensions of ensemble learning for mining non-stationary data streams can be found in the literature, most of them are crafted under a static base classifier and revisits preceding samples in the sliding window for a retraining step. This feature causes computationally prohibitive complexity and is not flexible enough to cope with rapidly changing environments. Their complexities are often demanding because it involves a large collection of offline classifiers due to the absence of structural complexities reduction mechanisms and lack of an online feature selection mechanism. A novel evolving ensemble classifier, namely Parsimonious Ensemble pENsemble, is proposed in this paper. pENsemble differs from existing architectures in the fact that it is built upon an evolving classifier from data streams, termed Parsimonious Classifier pClass. pENsemble is equipped by an ensemble pruning mechanism, which estimates a localized generalization error of a base classifier. A dynamic online feature selection scenario is integrated into the pENsemble. This method allows for dynamic selection and deselection of input features on the fly. pENsemble adopts a dynamic ensemble structure to output a final classification decision where it features a novel drift detection scenario to grow the ensemble structure. The efficacy of the pENsemble has been numerically demonstrated through rigorous numerical studies with dynamic and evolving data streams where it delivers the most encouraging performance in attaining a tradeoff between accuracy and complexity.
Tasks	Feature Selection
Published	2017-05-18
URL	https://arxiv.org/abs/1705.06460v2
PDF	https://arxiv.org/pdf/1705.06460v2.pdf
PWC	https://paperswithcode.com/paper/evolving-ensemble-fuzzy-classifier
Repo
Framework