Paper Group ANR 932
Left-Right Comparative Recurrent Model for Stereo Matching. Development and Evaluation of Recurrent Neural Network based Models for Hourly Traffic Volume and AADT Prediction. Open Set Adversarial Examples. Cahn–Hilliard inpainting with the double obstacle potential. AUEB at BioASQ 6: Document and Snippet Retrieval. Attending Category Disentangled …
Left-Right Comparative Recurrent Model for Stereo Matching
Title | Left-Right Comparative Recurrent Model for Stereo Matching |
Authors | Zequn Jie, Pengfei Wang, Yonggen Ling, Bo Zhao, Yunchao Wei, Jiashi Feng, Wei Liu |
Abstract | Leveraging the disparity information from both left and right views is crucial for stereo disparity estimation. Left-right consistency check is an effective way to enhance the disparity estimation by referring to the information from the opposite view. However, the conventional left-right consistency check is an isolated post-processing step and heavily hand-crafted. This paper proposes a novel left-right comparative recurrent model to perform left-right consistency checking jointly with disparity estimation. At each recurrent step, the model produces disparity results for both views, and then performs online left-right comparison to identify the mismatched regions which may probably contain erroneously labeled pixels. A soft attention mechanism is introduced, which employs the learned error maps for better guiding the model to selectively focus on refining the unreliable regions at the next recurrent step. In this way, the generated disparity maps are progressively improved by the proposed recurrent model. Extensive evaluations on KITTI 2015, Scene Flow and Middlebury benchmarks validate the effectiveness of our model, demonstrating that state-of-the-art stereo disparity estimation results can be achieved by this new model. |
Tasks | Disparity Estimation, Stereo Matching, Stereo Matching Hand |
Published | 2018-04-03 |
URL | http://arxiv.org/abs/1804.00796v1 |
http://arxiv.org/pdf/1804.00796v1.pdf | |
PWC | https://paperswithcode.com/paper/left-right-comparative-recurrent-model-for |
Repo | |
Framework | |
Development and Evaluation of Recurrent Neural Network based Models for Hourly Traffic Volume and AADT Prediction
Title | Development and Evaluation of Recurrent Neural Network based Models for Hourly Traffic Volume and AADT Prediction |
Authors | MD Zadid Khan, Sakib Mahmud Khan, Mashrur Chowdhury, Kakan Dey |
Abstract | The prediction of high-resolution hourly traffic volumes of a given roadway is essential for transportation planning. Traditionally, Automatic Traffic Recorders (ATR) are used to collect this hourly volume data. These large datasets are time series data characterized by long-term temporal dependencies and missing values. Regarding the temporal dependencies, all roadways are characterized by seasonal variations that can be weekly, monthly or yearly, depending on the cause of the variation. Regarding the missing data in a time-series sequence, traditional time series forecasting models perform poorly under the influence of seasonal variations. To address this limitation, robust, Recurrent Neural Network (RNN) based, multi-step ahead forecasting models are developed for time-series in this study. The simple RNN, the Gated Recurrent Unit (GRU) and the Long Short-Term Memory (LSTM) units are used to develop the model and evaluate its performance. Two approaches are used to address the missing value issue: masking and imputation, in conjunction with the RNN models. Six different imputation algorithms are then used to identify the best model. The analysis indicates that the LSTM model performs better than simple RNN and GRU models, and imputation performs better than masking to predict future traffic volume. Based on analysis using 92 ATRs, the LSTM-Median model is deemed the best model in all scenarios for hourly traffic volume and AADT prediction, with an average RMSE of 274 and MAPE of 18.91% for hourly traffic volume prediction and average RMSE of 824 and MAPE of 2.10% for AADT prediction. |
Tasks | Imputation, Time Series, Time Series Forecasting |
Published | 2018-08-15 |
URL | http://arxiv.org/abs/1808.10511v3 |
http://arxiv.org/pdf/1808.10511v3.pdf | |
PWC | https://paperswithcode.com/paper/development-and-evaluation-of-recurrent |
Repo | |
Framework | |
Open Set Adversarial Examples
Title | Open Set Adversarial Examples |
Authors | Zhedong Zheng, Liang Zheng, Zhilan Hu, Yi Yang |
Abstract | Adversarial examples in recent works target at closed set recognition systems, in which the training and testing classes are identical. In real-world scenarios, however, the testing classes may have limited, if any, overlap with the training classes, a problem named open set recognition. To our knowledge, the community does not have a specific design of adversarial examples targeting at this practical setting. Arguably, the new setting compromises traditional closed set attack methods in two aspects. First, closed set attack methods are based on classification and target at classification as well, but the open set problem suggests a different task, \emph{i.e.,} retrieval. It is undesirable that the generation mechanism of closed set recognition is different from the aim of open set recognition. Second, given that the query image is usually of an unseen class, predicting its category from the training classes is not reasonable, which leads to an inferior adversarial gradient. In this work, we view open set recognition as a retrieval task and propose a new approach, Opposite-Direction Feature Attack (ODFA), to generate adversarial examples / queries. When using an attacked example as query, we aim that the true matches be ranked as low as possible. In addressing the two limitations of closed set attack methods, ODFA directly works on the features for retrieval. The idea is to push away the feature of the adversarial query in the opposite direction of the original feature. Albeit simple, ODFA leads to a larger drop in Recall@K and mAP than the close-set attack methods on two open set recognition datasets, \emph{i.e.,} Market-1501 and CUB-200-2011. We also demonstrate that the attack performance of ODFA is not evidently superior to the state-of-the-art methods under closed set recognition (Cifar-10), suggesting its specificity for open set problems. |
Tasks | Open Set Learning |
Published | 2018-09-07 |
URL | http://arxiv.org/abs/1809.02681v1 |
http://arxiv.org/pdf/1809.02681v1.pdf | |
PWC | https://paperswithcode.com/paper/open-set-adversarial-examples |
Repo | |
Framework | |
Cahn–Hilliard inpainting with the double obstacle potential
Title | Cahn–Hilliard inpainting with the double obstacle potential |
Authors | Harald Garcke, Kei Fong Lam, Vanessa Styles |
Abstract | The inpainting of damaged images has a wide range of applications, and many different mathematical methods have been proposed to solve this problem. Inpainting with the help of Cahn–Hilliard models has been particularly successful, and it turns out that Cahn–Hilliard inpainting with the double obstacle potential can lead to better results compared to inpainting with a smooth double well potential. However, a mathematical analysis of this approach is missing so far. In this paper we give first analytical results for a Cahn–Hilliard double obstacle inpainting model regarding existence of global solutions to the time-dependent problem and stationary solutions to the time-independent problem without constraints on the parameters involved. With the help of numerical results we show the effectiveness of the approach for binary and grayscale images. |
Tasks | |
Published | 2018-01-17 |
URL | http://arxiv.org/abs/1801.05527v2 |
http://arxiv.org/pdf/1801.05527v2.pdf | |
PWC | https://paperswithcode.com/paper/cahn-hilliard-inpainting-with-the-double |
Repo | |
Framework | |
AUEB at BioASQ 6: Document and Snippet Retrieval
Title | AUEB at BioASQ 6: Document and Snippet Retrieval |
Authors | Georgios-Ioannis Brokos, Polyvios Liosis, Ryan McDonald, Dimitris Pappas, Ion Androutsopoulos |
Abstract | We present AUEB’s submissions to the BioASQ 6 document and snippet retrieval tasks (parts of Task 6b, Phase A). Our models use novel extensions to deep learning architectures that operate solely over the text of the query and candidate document/snippets. Our systems scored at the top or near the top for all batches of the challenge, highlighting the effectiveness of deep learning for these tasks. |
Tasks | |
Published | 2018-09-15 |
URL | http://arxiv.org/abs/1809.06366v1 |
http://arxiv.org/pdf/1809.06366v1.pdf | |
PWC | https://paperswithcode.com/paper/aueb-at-bioasq-6-document-and-snippet |
Repo | |
Framework | |
Attending Category Disentangled Global Context for Image Classification
Title | Attending Category Disentangled Global Context for Image Classification |
Authors | Keke Tang, Guodong Wei, Runnan Chen, Jie Zhu, Zhaoquan Gu, Wenping Wang |
Abstract | In this paper, we propose a general framework for image classification using the attention mechanism and global context, which could incorporate with various network architectures to improve their performance. To investigate the capability of the global context, we compare four mathematical models and observe the global context encoded in the category disentangled conditional generative model could give more guidance as “know what is task irrelevant will also know what is relevant”. Based on this observation, we define a novel Category Disentangled Global Context (CDGC) and devise a deep network to obtain it. By attending CDGC, the baseline networks could identify the objects of interest more accurately, thus improving the performance. We apply the framework to many different network architectures and compare with the state-of-the-art on four publicly available datasets. Extensive results validate the effectiveness and superiority of our approach. Code will be made public upon paper acceptance. |
Tasks | Image Classification |
Published | 2018-12-17 |
URL | https://arxiv.org/abs/1812.06663v4 |
https://arxiv.org/pdf/1812.06663v4.pdf | |
PWC | https://paperswithcode.com/paper/attending-category-disentangled-global |
Repo | |
Framework | |
The Monge-Kantorovich Optimal Transport Distance for Image Comparison
Title | The Monge-Kantorovich Optimal Transport Distance for Image Comparison |
Authors | Michael Snow, Jan Van lent |
Abstract | This paper focuses on the Monge-Kantorovich formulation of the optimal transport problem and the associated $L^2$ Wasserstein distance. We use the $L^2$ Wasserstein distance in the Nearest Neighbour (NN) machine learning architecture to demonstrate the potential power of the optimal transport distance for image comparison. We compare the Wasserstein distance to other established distances - including the partial differential equation (PDE) formulation of the optimal transport problem - and demonstrate that on the well known MNIST optical character recognition dataset, it achieves excellent results. |
Tasks | Optical Character Recognition |
Published | 2018-04-08 |
URL | http://arxiv.org/abs/1804.03531v1 |
http://arxiv.org/pdf/1804.03531v1.pdf | |
PWC | https://paperswithcode.com/paper/the-monge-kantorovich-optimal-transport |
Repo | |
Framework | |
Deep, Skinny Neural Networks are not Universal Approximators
Title | Deep, Skinny Neural Networks are not Universal Approximators |
Authors | Jesse Johnson |
Abstract | In order to choose a neural network architecture that will be effective for a particular modeling problem, one must understand the limitations imposed by each of the potential options. These limitations are typically described in terms of information theoretic bounds, or by comparing the relative complexity needed to approximate example functions between different architectures. In this paper, we examine the topological constraints that the architecture of a neural network imposes on the level sets of all the functions that it is able to approximate. This approach is novel for both the nature of the limitations and the fact that they are independent of network depth for a broad family of activation functions. |
Tasks | |
Published | 2018-09-30 |
URL | http://arxiv.org/abs/1810.00393v1 |
http://arxiv.org/pdf/1810.00393v1.pdf | |
PWC | https://paperswithcode.com/paper/deep-skinny-neural-networks-are-not-universal |
Repo | |
Framework | |
Subject2Vec: Generative-Discriminative Approach from a Set of Image Patches to a Vector
Title | Subject2Vec: Generative-Discriminative Approach from a Set of Image Patches to a Vector |
Authors | Sumedha Singla, Mingming Gong, Siamak Ravanbakhsh, Frank Sciurba, Barnabas Poczos, Kayhan N. Batmanghelich |
Abstract | We propose an attention-based method that aggregates local image features to a subject-level representation for predicting disease severity. In contrast to classical deep learning that requires a fixed dimensional input, our method operates on a set of image patches; hence it can accommodate variable length input image without image resizing. The model learns a clinically interpretable subject-level representation that is reflective of the disease severity. Our model consists of three mutually dependent modules which regulate each other: (1) a discriminative network that learns a fixed-length representation from local features and maps them to disease severity; (2) an attention mechanism that provides interpretability by focusing on the areas of the anatomy that contribute the most to the prediction task; and (3) a generative network that encourages the diversity of the local latent features. The generative term ensures that the attention weights are non-degenerate while maintaining the relevance of the local regions to the disease severity. We train our model end-to-end in the context of a large-scale lung CT study of Chronic Obstructive Pulmonary Disease (COPD). Our model gives state-of-the art performance in predicting clinical measures of severity for COPD. The distribution of the attention provides the regional relevance of lung tissue to the clinical measurements. |
Tasks | |
Published | 2018-06-28 |
URL | http://arxiv.org/abs/1806.11217v1 |
http://arxiv.org/pdf/1806.11217v1.pdf | |
PWC | https://paperswithcode.com/paper/subject2vec-generative-discriminative |
Repo | |
Framework | |
Max-Mahalanobis Linear Discriminant Analysis Networks
Title | Max-Mahalanobis Linear Discriminant Analysis Networks |
Authors | Tianyu Pang, Chao Du, Jun Zhu |
Abstract | A deep neural network (DNN) consists of a nonlinear transformation from an input to a feature representation, followed by a common softmax linear classifier. Though many efforts have been devoted to designing a proper architecture for nonlinear transformation, little investigation has been done on the classifier part. In this paper, we show that a properly designed classifier can improve robustness to adversarial attacks and lead to better prediction results. Specifically, we define a Max-Mahalanobis distribution (MMD) and theoretically show that if the input distributes as a MMD, the linear discriminant analysis (LDA) classifier will have the best robustness to adversarial examples. We further propose a novel Max-Mahalanobis linear discriminant analysis (MM-LDA) network, which explicitly maps a complicated data distribution in the input space to a MMD in the latent feature space and then applies LDA to make predictions. Our results demonstrate that the MM-LDA networks are significantly more robust to adversarial attacks, and have better performance in class-biased classification. |
Tasks | |
Published | 2018-02-26 |
URL | http://arxiv.org/abs/1802.09308v2 |
http://arxiv.org/pdf/1802.09308v2.pdf | |
PWC | https://paperswithcode.com/paper/max-mahalanobis-linear-discriminant-analysis |
Repo | |
Framework | |
Permutation Invariant Gaussian Matrix Models
Title | Permutation Invariant Gaussian Matrix Models |
Authors | Sanjaye Ramgoolam |
Abstract | Permutation invariant Gaussian matrix models were recently developed for applications in computational linguistics. A 5-parameter family of models was solved. In this paper, we use a representation theoretic approach to solve the general 13-parameter Gaussian model, which can be viewed as a zero-dimensional quantum field theory. We express the two linear and eleven quadratic terms in the action in terms of representation theoretic parameters. These parameters are coefficients of simple quadratic expressions in terms of appropriate linear combinations of the matrix variables transforming in specific irreducible representations of the symmetric group $S_D$ where $D$ is the size of the matrices. They allow the identification of constraints which ensure a convergent Gaussian measure and well-defined expectation values for polynomial functions of the random matrix at all orders. A graph-theoretic interpretation is known to allow the enumeration of permutation invariants of matrices at linear, quadratic and higher orders. We express the expectation values of all the quadratic graph-basis invariants and a selection of cubic and quartic invariants in terms of the representation theoretic parameters of the model. |
Tasks | |
Published | 2018-09-20 |
URL | https://arxiv.org/abs/1809.07559v2 |
https://arxiv.org/pdf/1809.07559v2.pdf | |
PWC | https://paperswithcode.com/paper/permutation-invariant-gaussian-matrix-models |
Repo | |
Framework | |
Revealing the Unobserved by Linking Collaborative Behavior and Side Knowledge
Title | Revealing the Unobserved by Linking Collaborative Behavior and Side Knowledge |
Authors | Evgeny Frolov, Ivan Oseledets |
Abstract | We propose a tensor-based model that fuses a more granular representation of user preferences with the ability to take additional side information into account. The model relies on the concept of ordinal nature of utility, which better corresponds to actual user perception. In addition to that, unlike the majority of hybrid recommenders, the model ties side information directly to collaborative data, which not only addresses the problem of extreme data sparsity, but also allows to naturally exploit patterns in the observed behavior for a more meaningful representation of user intents. We demonstrate the effectiveness of the proposed model on several standard benchmark datasets. The general formulation of the approach imposes no restrictions on the type of observed interactions and makes it potentially applicable for joint modelling of context information along with side data. |
Tasks | |
Published | 2018-07-27 |
URL | http://arxiv.org/abs/1807.10634v1 |
http://arxiv.org/pdf/1807.10634v1.pdf | |
PWC | https://paperswithcode.com/paper/revealing-the-unobserved-by-linking |
Repo | |
Framework | |
Inverse Transport Networks
Title | Inverse Transport Networks |
Authors | Chengqian Che, Fujun Luan, Shuang Zhao, Kavita Bala, Ioannis Gkioulekas |
Abstract | We introduce inverse transport networks as a learning architecture for inverse rendering problems where, given input image measurements, we seek to infer physical scene parameters such as shape, material, and illumination. During training, these networks are evaluated not only in terms of how close they can predict groundtruth parameters, but also in terms of whether the parameters they produce can be used, together with physically-accurate graphics renderers, to reproduce the input image measurements. To en- able training of inverse transport networks using stochastic gradient descent, we additionally create a general-purpose, physically-accurate differentiable renderer, which can be used to estimate derivatives of images with respect to arbitrary physical scene parameters. Our experiments demonstrate that inverse transport networks can be trained efficiently using differentiable rendering, and that they generalize to scenes with completely unseen geometry and illumination better than networks trained without appearance- matching regularization. |
Tasks | |
Published | 2018-09-28 |
URL | http://arxiv.org/abs/1809.10820v1 |
http://arxiv.org/pdf/1809.10820v1.pdf | |
PWC | https://paperswithcode.com/paper/inverse-transport-networks |
Repo | |
Framework | |
Single-view Object Shape Reconstruction Using Deep Shape Prior and Silhouette
Title | Single-view Object Shape Reconstruction Using Deep Shape Prior and Silhouette |
Authors | Kejie Li, Ravi Garg, Ming Cai, Ian Reid |
Abstract | 3D shape reconstruction from a single image is a highly ill-posed problem. Modern deep learning based systems try to solve this problem by learning an end-to-end mapping from image to shape via a deep network. In this paper, we aim to solve this problem via an online optimization framework inspired by traditional methods. Our framework employs a deep autoencoder to learn a set of latent codes of 3D object shapes, which are fitted by a probabilistic shape prior using Gaussian Mixture Model (GMM). At inference, the shape and pose are jointly optimized guided by both image cues and deep shape prior without relying on an initialization from any trained deep nets. Surprisingly, our method achieves comparable performance to state-of-the-art methods even without training an end-to-end network, which shows a promising step in this direction. |
Tasks | 3D Reconstruction, Object Reconstruction |
Published | 2018-11-29 |
URL | https://arxiv.org/abs/1811.11921v2 |
https://arxiv.org/pdf/1811.11921v2.pdf | |
PWC | https://paperswithcode.com/paper/optimizable-object-reconstruction-from-a |
Repo | |
Framework | |
A Cooperative Group Optimization System
Title | A Cooperative Group Optimization System |
Authors | Xiao-Feng Xie, Jiming Liu, Zun-Jing Wang |
Abstract | A cooperative group optimization (CGO) system is presented to implement CGO cases by integrating the advantages of the cooperative group and low-level algorithm portfolio design. Following the nature-inspired paradigm of a cooperative group, the agents not only explore in a parallel way with their individual memory, but also cooperate with their peers through the group memory. Each agent holds a portfolio of (heterogeneous) embedded search heuristics (ESHs), in which each ESH can drive the group into a stand-alone CGO case, and hybrid CGO cases in an algorithmic space can be defined by low-level cooperative search among a portfolio of ESHs through customized memory sharing. The optimization process might also be facilitated by a passive group leader through encoding knowledge in the search landscape. Based on a concrete framework, CGO cases are defined by a script assembling over instances of algorithmic components in a toolbox. A multilayer design of the script, with the support of the inherent updatable graph in the memory protocol, enables a simple way to address the challenge of accumulating heterogeneous ESHs and defining customized portfolios without any additional code. The CGO system is implemented for solving the constrained optimization problem with some generic components and only a few domain-specific components. Guided by the insights from algorithm portfolio design, customized CGO cases based on basic search operators can achieve competitive performance over existing algorithms as compared on a set of commonly-used benchmark instances. This work might provide a basic step toward a user-oriented development framework, since the algorithmic space might be easily evolved by accumulating competent ESHs. |
Tasks | |
Published | 2018-08-03 |
URL | http://arxiv.org/abs/1808.01342v1 |
http://arxiv.org/pdf/1808.01342v1.pdf | |
PWC | https://paperswithcode.com/paper/a-cooperative-group-optimization-system |
Repo | |
Framework | |