Paper Group ANR 686
A new measure of modularity density for community detection. Learning Rich Image Region Representation for Visual Question Answering. TDAPNet: Prototype Network with Recurrent Top-Down Attention for Robust Object Classification under Partial Occlusion. Overlapping community detection in networks based on link partitioning and partitioning around me …
A new measure of modularity density for community detection
| Title | A new measure of modularity density for community detection |
| Authors | Swathi M. Mula, Gerardo Veltri |
| Abstract | Using an intuitive concept of what constitutes a meaningful community, a novel metric is formulated for detecting non-overlapping communities in undirected, weighted heterogeneous networks. This metric, modularity density, is shown to be superior to the versions of modularity density in present literature. Compared to the previous versions of modularity density, maximization of our metric is proven to be free from bias and better detect weakly-separated communities particularly in heterogeneous networks. In addition to these characteristics, the computational running time of our modularity density is found to be on par or faster than that of the previous variants. Our findings further reveal that community detection by maximization of our metric is mathematically related to partitioning a network by minimization of the normalized cut criterion. |
| Tasks | Community Detection |
| Published | 2019-08-22 |
| URL | https://arxiv.org/abs/1908.08452v1 |
| https://arxiv.org/pdf/1908.08452v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-new-measure-of-modularity-density-for |
| Repo | |
| Framework | |
Learning Rich Image Region Representation for Visual Question Answering
| Title | Learning Rich Image Region Representation for Visual Question Answering |
| Authors | Bei Liu, Zhicheng Huang, Zhaoyang Zeng, Zheyu Chen, Jianlong Fu |
| Abstract | We propose to boost VQA by leveraging more powerful feature extractors by improving the representation ability of both visual and text features and the ensemble of models. For visual feature, some detection techniques are used to improve the detector. For text feature, we adopt BERT as the language model and find that it can significantly improve VQA performance. Our solution won the second place in the VQA Challenge 2019. |
| Tasks | |
| Published | 2019-10-29 |
| URL | https://arxiv.org/abs/1910.13077v1 |
| https://arxiv.org/pdf/1910.13077v1.pdf | |
| PWC | https://paperswithcode.com/paper/learning-rich-image-region-representation-for |
| Repo | |
| Framework | |
TDAPNet: Prototype Network with Recurrent Top-Down Attention for Robust Object Classification under Partial Occlusion
| Title | TDAPNet: Prototype Network with Recurrent Top-Down Attention for Robust Object Classification under Partial Occlusion |
| Authors | Mingqing Xiao, Adam Kortylewski, Ruihai Wu, Siyuan Qiao, Wei Shen, Alan Yuille |
| Abstract | Despite deep convolutional neural networks’ great success in object classification, it suffers from severe generalization performance drop under occlusion due to the inconsistency between training and testing data. Because of the large variance of occluders, our goal is a model trained on occlusion-free data while generalizable to occlusion conditions. In this work, we integrate prototypes, partial matching and top-down attention regulation into deep neural networks to realize robust object classification under occlusion. We first introduce prototype learning as its regularization encourages compact data clusters, which enables better generalization ability under inconsistent conditions. Then, attention map at intermediate layer based on feature dictionary and activation scale is estimated for partial matching, which sifts irrelevant information out when comparing features with prototypes. Further, inspired by neuroscience research that reveals the important role of feedback connection for object recognition under occlusion, a top-down feedback attention regulation is introduced into convolution layers, purposefully reducing the contamination by occlusion during feature extraction stage. Our experiment results on partially occluded MNIST and vehicles from the PASCAL3D+ dataset demonstrate that the proposed network significantly improves the robustness of current deep neural networks under occlusion. Our code will be released. |
| Tasks | Object Classification, Object Recognition |
| Published | 2019-09-09 |
| URL | https://arxiv.org/abs/1909.03879v2 |
| https://arxiv.org/pdf/1909.03879v2.pdf | |
| PWC | https://paperswithcode.com/paper/tdapnet-prototype-network-with-recurrent-top |
| Repo | |
| Framework | |
Overlapping community detection in networks based on link partitioning and partitioning around medoids
| Title | Overlapping community detection in networks based on link partitioning and partitioning around medoids |
| Authors | Alexander Ponomarenko, Leonidas Pitsoulis, Marat Shamshetdinov |
| Abstract | In this paper, we present a new method for detecting overlapping communities in networks with a predefined number of clusters. The overlapping communities in the graph are obtained by detecting the disjoint communities in the associated line graph by means of link partitioning and partitioning around medoids. Partitioning around medoids is done through the use of a distance function defined on the set of nodes of the linear graph. In the present paper, we consider the commute distance and amplified commute distance functions as distance functions. The performance of the proposed method is demonstrated by computational experiments on real-life instances. |
| Tasks | Community Detection |
| Published | 2019-07-20 |
| URL | https://arxiv.org/abs/1907.08731v1 |
| https://arxiv.org/pdf/1907.08731v1.pdf | |
| PWC | https://paperswithcode.com/paper/overlapping-community-detection-in-networks |
| Repo | |
| Framework | |
Modelling bid-ask spread conditional distributions using hierarchical correlation reconstruction
| Title | Modelling bid-ask spread conditional distributions using hierarchical correlation reconstruction |
| Authors | Jarosław Duda, Robert Syrek, Henryk Gurgul |
| Abstract | While we would like to predict exact values, available incomplete information is rarely sufficient - usually allowing only to predict conditional probability distributions. This article discusses hierarchical correlation reconstruction (HCR) methodology for such prediction on example of usually unavailable bid-ask spreads, predicted from more accessible data like closing price, volume, high/low price, returns. In HCR methodology we first normalize marginal distributions to nearly uniform like in copula theory. Then we model (joint) densities as linear combinations of orthonormal polynomials, getting its decomposition into (mixed) moments. Then here we model each moment (separately) of predicted variable as a linear combination of mixed moments of known variables using least squares linear regression - getting accurate description with interpretable coefficients describing linear relations between moments. Combining such predicted moments we get predicted density as a polynomial, for which we can e.g. calculate expected value, but also variance to evaluate uncertainty of such prediction, or we can use the entire distribution e.g. for more accurate further calculations or generating random values. There were performed 10-fold cross-validation log-likelihood tests for 22 DAX companies, leading to very accurate predictions, especially when using individual models for each company as there were found large differences between their behaviors. Additional advantage of the discussed methodology is being computationally inexpensive, finding and evaluation a model with hundreds of parameters and thousands of data points takes a second on a laptop. |
| Tasks | |
| Published | 2019-11-04 |
| URL | https://arxiv.org/abs/1911.02361v1 |
| https://arxiv.org/pdf/1911.02361v1.pdf | |
| PWC | https://paperswithcode.com/paper/modelling-bid-ask-spread-conditional |
| Repo | |
| Framework | |
Curate and Generate: A Corpus and Method for Joint Control of Semantics and Style in Neural NLG
| Title | Curate and Generate: A Corpus and Method for Joint Control of Semantics and Style in Neural NLG |
| Authors | Shereen Oraby, Vrindavan Harrison, Abteen Ebrahimi, Marilyn Walker |
| Abstract | Neural natural language generation (NNLG) from structured meaning representations has become increasingly popular in recent years. While we have seen progress with generating syntactically correct utterances that preserve semantics, various shortcomings of NNLG systems are clear: new tasks require new training data which is not available or straightforward to acquire, and model outputs are simple and may be dull and repetitive. This paper addresses these two critical challenges in NNLG by: (1) scalably (and at no cost) creating training datasets of parallel meaning representations and reference texts with rich style markup by using data from freely available and naturally descriptive user reviews, and (2) systematically exploring how the style markup enables joint control of semantic and stylistic aspects of neural model output. We present YelpNLG, a corpus of 300,000 rich, parallel meaning representations and highly stylistically varied reference texts spanning different restaurant attributes, and describe a novel methodology that can be scalably reused to generate NLG datasets for other domains. The experiments show that the models control important aspects, including lexical choice of adjectives, output length, and sentiment, allowing the models to successfully hit multiple style targets without sacrificing semantics. |
| Tasks | Text Generation |
| Published | 2019-06-04 |
| URL | https://arxiv.org/abs/1906.01334v2 |
| https://arxiv.org/pdf/1906.01334v2.pdf | |
| PWC | https://paperswithcode.com/paper/curate-and-generate-a-corpus-and-method-for |
| Repo | |
| Framework | |
What Can Neural Networks Reason About?
| Title | What Can Neural Networks Reason About? |
| Authors | Keyulu Xu, Jingling Li, Mozhi Zhang, Simon S. Du, Ken-ichi Kawarabayashi, Stefanie Jegelka |
| Abstract | Neural networks have succeeded in many reasoning tasks. Empirically, these tasks require specialized network structures, e.g., Graph Neural Networks (GNNs) perform well on many such tasks, but less structured networks fail. Theoretically, there is limited understanding of why and when a network structure generalizes better than others, although they have equal expressive power. In this paper, we develop a framework to characterize which reasoning tasks a network can learn well, by studying how well its computation structure aligns with the algorithmic structure of the relevant reasoning process. We formally define this algorithmic alignment and derive a sample complexity bound that decreases with better alignment. This framework offers an explanation for the empirical success of popular reasoning models, and suggests their limitations. As an example, we unify seemingly different reasoning tasks, such as intuitive physics, visual question answering, and shortest paths, via the lens of a powerful algorithmic paradigm, dynamic programming (DP). We show that GNNs align with DP and thus are expected to solve these tasks. On several reasoning tasks, our theory is supported by empirical results. |
| Tasks | Question Answering, Visual Question Answering |
| Published | 2019-05-30 |
| URL | https://arxiv.org/abs/1905.13211v4 |
| https://arxiv.org/pdf/1905.13211v4.pdf | |
| PWC | https://paperswithcode.com/paper/what-can-neural-networks-reason-about |
| Repo | |
| Framework | |
Character-based NMT with Transformer
| Title | Character-based NMT with Transformer |
| Authors | Rohit Gupta, Laurent Besacier, Marc Dymetman, Matthias Gallé |
| Abstract | Character-based translation has several appealing advantages, but its performance is in general worse than a carefully tuned BPE baseline. In this paper we study the impact of character-based input and output with the Transformer architecture. In particular, our experiments on EN-DE show that character-based Transformer models are more robust than their BPE counterpart, both when translating noisy text, and when translating text from a different domain. To obtain comparable BLEU scores in clean, in-domain data and close the gap with BPE-based models we use known techniques to train deeper Transformer models. |
| Tasks | |
| Published | 2019-11-12 |
| URL | https://arxiv.org/abs/1911.04997v1 |
| https://arxiv.org/pdf/1911.04997v1.pdf | |
| PWC | https://paperswithcode.com/paper/character-based-nmt-with-transformer |
| Repo | |
| Framework | |
More Supervision, Less Computation: Statistical-Computational Tradeoffs in Weakly Supervised Learning
| Title | More Supervision, Less Computation: Statistical-Computational Tradeoffs in Weakly Supervised Learning |
| Authors | Xinyang Yi, Zhaoran Wang, Zhuoran Yang, Constantine Caramanis, Han Liu |
| Abstract | We consider the weakly supervised binary classification problem where the labels are randomly flipped with probability $1- {\alpha}$. Although there exist numerous algorithms for this problem, it remains theoretically unexplored how the statistical accuracies and computational efficiency of these algorithms depend on the degree of supervision, which is quantified by ${\alpha}$. In this paper, we characterize the effect of ${\alpha}$ by establishing the information-theoretic and computational boundaries, namely, the minimax-optimal statistical accuracy that can be achieved by all algorithms, and polynomial-time algorithms under an oracle computational model. For small ${\alpha}$, our result shows a gap between these two boundaries, which represents the computational price of achieving the information-theoretic boundary due to the lack of supervision. Interestingly, we also show that this gap narrows as ${\alpha}$ increases. In other words, having more supervision, i.e., more correct labels, not only improves the optimal statistical accuracy as expected, but also enhances the computational efficiency for achieving such accuracy. |
| Tasks | |
| Published | 2019-07-14 |
| URL | https://arxiv.org/abs/1907.06257v1 |
| https://arxiv.org/pdf/1907.06257v1.pdf | |
| PWC | https://paperswithcode.com/paper/more-supervision-less-computation-statistical-1 |
| Repo | |
| Framework | |
End-To-End Measure for Text Recognition
| Title | End-To-End Measure for Text Recognition |
| Authors | Gundram Leifert, Roger Labahn, Tobias Grüning, Svenja Leifert |
| Abstract | Measuring the performance of text recognition and text line detection engines is an important step to objectively compare systems and their configuration. There exist well-established measures for both tasks separately. However, there is no sophisticated evaluation scheme to measure the quality of a combined text line detection and text recognition system. The F-measure on word level is a well-known methodology, which is sometimes used in this context. Nevertheless, it does not take into account the alignment of hypothesis and ground truth text and can lead to deceptive results. Since users of automatic information retrieval pipelines in the context of text recognition are mainly interested in the end-to-end performance of a given system, there is a strong need for such a measure. Hence, we present a measure to evaluate the quality of an end-to-end text recognition system. The basis for this measure is the well established and widely used character error rate, which is limited – in its original form – to aligned hypothesis and ground truth texts. The proposed measure is flexible in a way that it can be configured to penalize different reading orders between the hypothesis and ground truth and can take into account the geometric position of the text lines. Additionally, it can ignore over- and under- segmentation of text lines. With these parameters it is possible to get a measure fitting best to its own needs. |
| Tasks | Information Retrieval |
| Published | 2019-08-26 |
| URL | https://arxiv.org/abs/1908.09584v1 |
| https://arxiv.org/pdf/1908.09584v1.pdf | |
| PWC | https://paperswithcode.com/paper/end-to-end-measure-for-text-recognition |
| Repo | |
| Framework | |
The State of NLP Literature: A Diachronic Analysis of the ACL Anthology
| Title | The State of NLP Literature: A Diachronic Analysis of the ACL Anthology |
| Authors | Saif M. Mohammad |
| Abstract | The ACL Anthology (AA) is a digital repository of tens of thousands of articles on Natural Language Processing (NLP). This paper examines the literature as a whole to identify broad trends in productivity, focus, and impact. It presents the analyses in a sequence of questions and answers. The goal is to record the state of the AA literature: who and how many of us are publishing? what are we publishing on? where and in what form are we publishing? and what is the impact of our publications? The answers are usually in the form of numbers, graphs, and inter-connected visualizations. Special emphasis is laid on the demographics and inclusiveness of NLP publishing. Notably, we find that only about 30% of first authors are female, and that this percentage has not improved since the year 2000. We also show that, on average, female first authors are cited less than male first authors, even when controlling for experience. We hope that recording citation and participation gaps across demographic groups will encourage more inclusiveness and fairness in research. |
| Tasks | |
| Published | 2019-11-08 |
| URL | https://arxiv.org/abs/1911.03562v1 |
| https://arxiv.org/pdf/1911.03562v1.pdf | |
| PWC | https://paperswithcode.com/paper/the-state-of-nlp-literature-a-diachronic |
| Repo | |
| Framework | |
Embedding-based Silhouette Community Detection
| Title | Embedding-based Silhouette Community Detection |
| Authors | Blaž Škrlj, Jan Kralj, Nada Lavrač |
| Abstract | Mining complex data in the form of networks is of increasing interest in many scientific disciplines. Network communities correspond to densely connected subnetworks, and often represent key functional parts of real-world systems. In this work, we propose Silhouette Community Detection (SCD), an approach for detecting communities, based on clustering of network node embeddings, i.e. real valued representations of nodes derived from their neighborhoods. We investigate the performance of the proposed SCD approach on 234 synthetic networks, as well as on a real-life social network. Even though SCD is not based on any form of modularity optimization, it performs comparably or better than state-of-the-art community detection algorithms, such as the InfoMap and Louvain algorithms. Further, we demonstrate how SCD’s outputs can be used along with domain ontologies in semantic subgroup discovery, yielding human-understandable explanations of communities detected in a real-life protein interaction network. Being embedding-based, SCD is widely applicable and can be tested out-of-the-box as part of many existing network learning and exploration pipelines. |
| Tasks | Community Detection |
| Published | 2019-07-17 |
| URL | https://arxiv.org/abs/1908.02556v1 |
| https://arxiv.org/pdf/1908.02556v1.pdf | |
| PWC | https://paperswithcode.com/paper/embedding-based-silhouette-community |
| Repo | |
| Framework | |
Deformable Registration Using Average Geometric Transformations for Brain MR Images
| Title | Deformable Registration Using Average Geometric Transformations for Brain MR Images |
| Authors | Yongpei Zhu, Zicong Zhou, Guojun Liao, Kehong Yuan |
| Abstract | Accurate registration of medical images is vital for doctor’s diagnosis and quantitative analysis. In this paper, we propose a new deformable medical image registration method based on average geometric transformations and VoxelMorph CNN architecture. We compute the differential geometric information including Jacobian determinant(JD) and the curl vector(CV) of diffeomorphic registration field and use them as multi-channel of VoxelMorph CNN for second train. In addition, we use the average transformation to construct a standard brain MRI atlas which can be used as fixed image. We verify our method on two datasets including ADNI dataset and MRBrainS18 Challenge dataset, and obtain excellent improvement on MR image registration with average Dice scores and non-negative Jacobian locations compared with MIT’s original method. The experimental results show the method can achieve better performance in brain MRI diagnosis. |
| Tasks | Deformable Medical Image Registration, Image Registration, Medical Image Registration |
| Published | 2019-07-23 |
| URL | https://arxiv.org/abs/1907.09670v1 |
| https://arxiv.org/pdf/1907.09670v1.pdf | |
| PWC | https://paperswithcode.com/paper/deformable-registration-using-average |
| Repo | |
| Framework | |
Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation
| Title | Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation |
| Authors | Juncheng Li, Xin Wang, Siliang Tang, Haizhou Shi, Fei Wu, Yueting Zhuang, William Yang Wang |
| Abstract | Visual navigation is a task of training an embodied agent by intelligently navigating to a target object (e.g., television) using only visual observations. A key challenge for current deep reinforcement learning models lies in the requirements for a large amount of training data. It is exceedingly expensive to construct sufficient 3D synthetic environments annotated with the target object information. In this paper, we focus on visual navigation in the low-resource setting, where we have only a few training environments annotated with object information. We propose a novel unsupervised reinforcement learning approach to learn transferable meta-skills (e.g., bypass obstacles, go straight) from unannotated environments without any supervisory signals. The agent can then fast adapt to visual navigation through learning a high-level master policy to combine these meta-skills, when the visual-navigation-specified reward is provided. Evaluation in the AI2-THOR environments shows that our method significantly outperforms the baseline by 53.34% relatively on SPL, and further qualitative analysis demonstrates that our method learns transferable motor primitives for visual navigation. |
| Tasks | Visual Navigation |
| Published | 2019-11-18 |
| URL | https://arxiv.org/abs/1911.07450v1 |
| https://arxiv.org/pdf/1911.07450v1.pdf | |
| PWC | https://paperswithcode.com/paper/unsupervised-reinforcement-learning-of |
| Repo | |
| Framework | |
Hierarchical Clustering Supported by Reciprocal Nearest Neighbors
| Title | Hierarchical Clustering Supported by Reciprocal Nearest Neighbors |
| Authors | Wen-Bo Xie, Yan-Li Lee, Cong Wang, Duan-Bing Chen, Tao Zhou |
| Abstract | Clustering is a fundamental analysis tool aiming at classifying data points into groups based on their similarity or distance. It has found successful applications in all natural and social sciences, including biology, physics, economics, chemistry, astronomy, psychology, and so on. Among numerous existent algorithms, hierarchical clustering algorithms are of a particular advantage as they can provide results under different resolutions without any predetermined number of clusters and unfold the organization of resulted clusters. At the same time, they suffer a variety of drawbacks and thus are either time-consuming or inaccurate. We propose a novel hierarchical clustering approach on the basis of a simple hypothesis that two reciprocal nearest data points should be grouped in one cluster. Extensive tests on data sets across multiple domains show that our method is much faster and more accurate than the state-of-the-art benchmarks. We further extend our method to deal with the community detection problem in real networks, achieving remarkably better results in comparison with the well-known Girvan-Newman algorithm. |
| Tasks | Community Detection |
| Published | 2019-07-09 |
| URL | https://arxiv.org/abs/1907.04915v1 |
| https://arxiv.org/pdf/1907.04915v1.pdf | |
| PWC | https://paperswithcode.com/paper/hierarchical-clustering-supported-by |
| Repo | |
| Framework | |
