January 26, 2020

3330 words 16 mins read

Paper Group ANR 1383

Towards Optimisation of Collaborative Question Answering over Knowledge Graphs. A Novel Initial Clusters Generation Method for K-means-based Clustering Algorithms for Mixed Datasets. Answering Complex Questions by Joining Multi-Document Evidence with Quasi Knowledge Graphs. Dense Fusion Classmate Network for Land Cover Classification. Improve Diver …

Towards Optimisation of Collaborative Question Answering over Knowledge Graphs


Title	Towards Optimisation of Collaborative Question Answering over Knowledge Graphs
Authors	Kuldeep Singh, Mohamad Yaser Jaradeh, Saeedeh Shekarpour, Akash Kulkarni, Arun Sethupat Radhakrishna, Ioanna Lytra, Maria-Esther Vidal, Jens Lehmann
Abstract	Collaborative Question Answering (CQA) frameworks for knowledge graphs aim at integrating existing question answering (QA) components for implementing sequences of QA tasks (i.e. QA pipelines). The research community has paid substantial attention to CQAs since they support reusability and scalability of the available components in addition to the flexibility of pipelines. CQA frameworks attempt to build such pipelines automatically by solving two optimisation problems: 1) local collective performance of QA components per QA task and 2) global performance of QA pipelines. In spite offering several advantages over monolithic QA systems, the effectiveness and efficiency of CQA frameworks in answering questions is limited. In this paper, we tackle the problem of local optimisation of CQA frameworks and propose a three fold approach, which applies feature selection techniques with supervised machine learning approaches in order to identify the best performing components efficiently. We have empirically evaluated our approach over existing benchmarks and compared to existing automatic CQA frameworks. The observed results provide evidence that our approach answers a higher number of questions than the state of the art while reducing: i) the number of used features by 50% and ii) the number of components used by 76%.
Tasks	Feature Selection, Knowledge Graphs, Question Answering
Published	2019-08-14
URL	https://arxiv.org/abs/1908.05098v1
PDF	https://arxiv.org/pdf/1908.05098v1.pdf
PWC	https://paperswithcode.com/paper/towards-optimisation-of-collaborative
Repo
Framework

A Novel Initial Clusters Generation Method for K-means-based Clustering Algorithms for Mixed Datasets


Title	A Novel Initial Clusters Generation Method for K-means-based Clustering Algorithms for Mixed Datasets
Authors	Amir Ahmad, Shehroz S. Khan
Abstract	Mixed datasets consist of both numeric and categorical attributes. Various K-means-based clustering algorithms have been developed to cluster these datasets. Generally, these algorithms use random partition as a starting point, which tend to produce different clustering results in different runs. This inconsistency of clustering results may lead to unreliable inferences from the data. A few initialization algorithms have been developed to compute initial partition for mixed datasets; however, they are either computationally expensive or they do not produce consistent clustering results in different runs. In this paper, we propose, initKmix, a novel approach to find initial partition for K-means-based clustering algorithms for mixed datasets. The initKmix is based on the experimental observations that (i) some data points in a dataset remain in the same clusters created by k-means-based clustering algorithm irrespective of the choice of initial clusters, and (ii) individual attribute information can be used to create initial clusters. In initKmix method, a k-means-based clustering algorithm is run many times, in each run one of the attribute is used to produce initial partition. The clustering results of various runs are combined to produce initial partition. This initial partition is then be used as a seed to a k-means-based clustering algorithm to cluster mixed data. The initial partitions produced by initKmix are always fixed, do not change over different runs or by changing the order of the data objects. Experiments with various categorical and mixed datasets showed that initKmix produced accurate and consistent results, and outperformed random initialization and other state-of-the-art initialization methods. Experiments also showed that K-means-based clustering for mixed datasets with initKmix outperformed many state-of-the-art clustering algorithms.
Tasks
Published	2019-01-31
URL	http://arxiv.org/abs/1902.00127v2
PDF	http://arxiv.org/pdf/1902.00127v2.pdf
PWC	https://paperswithcode.com/paper/a-novel-initial-clusters-generation-method
Repo
Framework

Answering Complex Questions by Joining Multi-Document Evidence with Quasi Knowledge Graphs


Title	Answering Complex Questions by Joining Multi-Document Evidence with Quasi Knowledge Graphs
Authors	Xiaolu Lu, Soumajit Pramanik, Rishiraj Saha Roy, Abdalghani Abujabal, Yafang Wang, Gerhard Weikum
Abstract	Direct answering of questions that involve multiple entities and relations is a challenge for text-based QA. This problem is most pronounced when answers can be found only by joining evidence from multiple documents. Curated knowledge graphs (KGs) may yield good answers, but are limited by their inherent incompleteness and potential staleness. This paper presents QUEST, a method that can answer complex questions directly from textual sources on-the-fly, by computing similarity joins over partial results from different documents. Our method is completely unsupervised, avoiding training-data bottlenecks and being able to cope with rapidly evolving ad hoc topics and formulation style in user questions. QUEST builds a noisy quasi KG with node and edge weights, consisting of dynamically retrieved entity names and relational phrases. It augments this graph with types and semantic alignments, and computes the best answers by an algorithm for Group Steiner Trees. We evaluate QUEST on benchmarks of complex questions, and show that it substantially outperforms state-of-the-art baselines.
Tasks	Knowledge Graphs
Published	2019-08-01
URL	https://arxiv.org/abs/1908.00469v2
PDF	https://arxiv.org/pdf/1908.00469v2.pdf
PWC	https://paperswithcode.com/paper/answering-complex-questions-by-joining-multi
Repo
Framework

Dense Fusion Classmate Network for Land Cover Classification


Title	Dense Fusion Classmate Network for Land Cover Classification
Authors	Chao Tian, Cong Li, Jianping Shi
Abstract	Recently, FCNs based methods have made great progress in semantic segmentation. Different with ordinary scenes, satellite image owns specific characteristics, which elements always extend to large scope and no regular or clear boundaries. Therefore, effective mid-level structure information extremely missing, precise pixel-level classification becomes tough issues. In this paper, a Dense Fusion Classmate Network (DFCNet) is proposed to adopt in land cover classification.
Tasks	Semantic Segmentation
Published	2019-11-19
URL	https://arxiv.org/abs/1911.08169v1
PDF	https://arxiv.org/pdf/1911.08169v1.pdf
PWC	https://paperswithcode.com/paper/dense-fusion-classmate-network-for-land-cover
Repo
Framework

Improve Diverse Text Generation by Self Labeling Conditional Variational Auto Encoder


Title	Improve Diverse Text Generation by Self Labeling Conditional Variational Auto Encoder
Authors	Yuchi Zhang, Yongliang Wang, Liping Zhang, Zhiqiang Zhang, Kun Gai
Abstract	Diversity plays a vital role in many text generating applications. In recent years, Conditional Variational Auto Encoders (CVAE) have shown promising performances for this task. However, they often encounter the so called KL-Vanishing problem. Previous works mitigated such problem by heuristic methods such as strengthening the encoder or weakening the decoder while optimizing the CVAE objective function. Nevertheless, the optimizing direction of these methods are implicit and it is hard to find an appropriate degree to which these methods should be applied. In this paper, we propose an explicit optimizing objective to complement the CVAE to directly pull away from KL-vanishing. In fact, this objective term guides the encoder towards the “best encoder” of the decoder to enhance the expressiveness. A labeling network is introduced to estimate the “best encoder”. It provides a continuous label in the latent space of CVAE to help build a close connection between latent variables and targets. The whole proposed method is named Self Labeling CVAE~(SLCVAE). To accelerate the research of diverse text generation, we also propose a large native one-to-many dataset. Extensive experiments are conducted on two tasks, which show that our method largely improves the generating diversity while achieving comparable accuracy compared with state-of-art algorithms.
Tasks	Text Generation
Published	2019-03-26
URL	http://arxiv.org/abs/1903.10842v1
PDF	http://arxiv.org/pdf/1903.10842v1.pdf
PWC	https://paperswithcode.com/paper/improve-diverse-text-generation-by-self
Repo
Framework

Improvability Through Semi-Supervised Learning: A Survey of Theoretical Results


Title	Improvability Through Semi-Supervised Learning: A Survey of Theoretical Results
Authors	Alexander Mey, Marco Loog
Abstract	Semi-supervised learning is a setting in which one has labeled and unlabeled data available. In this survey we explore different types of theoretical results when one uses unlabeled data in classification and regression tasks. Most methods that use unlabeled data rely on certain assumptions about the data distribution. When those assumptions are not met in reality, including unlabeled data may actually decrease performance. Studying such methods, it therefore is particularly important to have an understanding of the underlying theory. In this review we gather results about the possible gains one can achieve when using semi-supervised learning as well as results about the limits of such methods. More precisely, this review collects the answers to the following questions: What are, in terms of improving supervised methods, the limits of semi-supervised learning? What are the assumptions of different methods? What can we achieve if the assumptions are true? Finally, we also discuss the biggest bottleneck of semi-supervised learning, namely the assumptions they make.
Tasks
Published	2019-08-26
URL	https://arxiv.org/abs/1908.09574v2
PDF	https://arxiv.org/pdf/1908.09574v2.pdf
PWC	https://paperswithcode.com/paper/improvability-through-semi-supervised
Repo
Framework

A Neural Turing~Machine for Conditional Transition Graph Modeling


Title	A Neural Turing~Machine for Conditional Transition Graph Modeling
Authors	Mehdi Ben Lazreg, Morten Goodwin, Ole-Christoffer Granmo
Abstract	Graphs are an essential part of many machine learning problems such as analysis of parse trees, social networks, knowledge graphs, transportation systems, and molecular structures. Applying machine learning in these areas typically involves learning the graph structure and the relationship between the nodes of the graph. However, learning the graph structure is often complex, particularly when the graph is cyclic, and the transitions from one node to another are conditioned such as graphs used to represent a finite state machine. To solve this problem, we propose to extend the memory based Neural Turing Machine (NTM) with two novel additions. We allow for transitions between nodes to be influenced by information received from external environments, and we let the NTM learn the context of those transitions. We refer to this extension as the Conditional Neural Turing Machine (CNTM). We show that the CNTM can infer conditional transition graphs by empirically verifiying the model on two data sets: a large set of randomly generated graphs, and a graph modeling the information retrieval process during certain crisis situations. The results show that the CNTM is able to reproduce the paths inside the graph with accuracy ranging from 82,12% for 10 nodes graphs to 65,25% for 100 nodes graphs.
Tasks	Information Retrieval, Knowledge Graphs
Published	2019-07-15
URL	https://arxiv.org/abs/1907.06432v1
PDF	https://arxiv.org/pdf/1907.06432v1.pdf
PWC	https://paperswithcode.com/paper/a-neural-turingmachine-for-conditional
Repo
Framework

Cross Domain Knowledge Transfer for Unsupervised Vehicle Re-identification


Title	Cross Domain Knowledge Transfer for Unsupervised Vehicle Re-identification
Authors	Jinjia Peng, Huibing Wang, Tongtong Zhao, Xianping Fu
Abstract	Vehicle re-identification (reID) is to identify a target vehicle in different cameras with non-overlapping views. When deploy the well-trained model to a new dataset directly, there is a severe performance drop because of differences among datasets named domain bias. To address this problem, this paper proposes an domain adaptation framework which contains an image-to-image translation network named vehicle transfer generative adversarial network (VTGAN) and an attention-based feature learning network (ATTNet). VTGAN could make images from the source domain (well-labeled) have the style of target domain (unlabeled) and preserve identity information of source domain. To further improve the domain adaptation ability for various backgrounds, ATTNet is proposed to train generated images with the attention structure for vehicle reID. Comprehensive experimental results clearly demonstrate that our method achieves excellent performance on VehicleID dataset.
Tasks	Domain Adaptation, Image-to-Image Translation, Transfer Learning, Vehicle Re-Identification
Published	2019-03-19
URL	http://arxiv.org/abs/1903.07868v1
PDF	http://arxiv.org/pdf/1903.07868v1.pdf
PWC	https://paperswithcode.com/paper/cross-domain-knowledge-transfer-for
Repo
Framework

Mutual Reinforcement Learning


Title	Mutual Reinforcement Learning
Authors	Sayanti Roy, Emily Kieson, Charles Abramson, Christopher Crick
Abstract	Recently, collaborative robots have begun to train humans to achieve complex tasks, and the mutual information exchange between them can lead to successful robot-human collaborations. In this paper we demonstrate the application and effectiveness of a new approach called mutual reinforcement learning (MRL), where both humans and autonomous agents act as reinforcement learners in a skill transfer scenario over continuous communication and feedback. An autonomous agent initially acts as an instructor who can teach a novice human participant complex skills using the MRL strategy. While teaching skills in a physical (block-building) ($n=34$) or simulated (Tetris) environment ($n=31$), the expert tries to identify appropriate reward channels preferred by each individual and adapts itself accordingly using an exploration-exploitation strategy. These reward channel preferences can identify important behaviors of the human participants, because they may well exercise the same behaviors in similar situations later. In this way, skill transfer takes place between an expert system and a novice human operator. We divided the subject population into three groups and observed the skill transfer phenomenon, analyzing it with Simpson"s psychometric model. 5-point Likert scales were also used to identify the cognitive models of the human participants. We obtained a shared cognitive model which not only improves human cognition but enhances the robot’s cognitive strategy to understand the mental model of its human partners while building a successful robot-human collaborative framework.
Tasks
Published	2019-07-15
URL	https://arxiv.org/abs/1907.06725v3
PDF	https://arxiv.org/pdf/1907.06725v3.pdf
PWC	https://paperswithcode.com/paper/mutual-reinforcement-learning
Repo
Framework

Automatic Segmentation of Pulmonary Lobes Using a Progressive Dense V-Network


Title	Automatic Segmentation of Pulmonary Lobes Using a Progressive Dense V-Network
Authors	Abdullah-Al-Zubaer Imran, Ali Hatamizadeh, Shilpa P. Ananth, Xiaowei Ding, Demetri Terzopoulos, Nima Tajbakhsh
Abstract	Reliable and automatic segmentation of lung lobes is important for diagnosis, assessment, and quantification of pulmonary diseases. The existing techniques are prohibitively slow, undesirably rely on prior (airway/vessel) segmentation, and/or require user interactions for optimal results. This work presents a reliable, fast, and fully automated lung lobe segmentation based on a progressive dense V-network (PDV-Net). The proposed method can segment lung lobes in one forward pass of the network, with an average runtime of 2 seconds using 1 Nvidia Titan XP GPU, eliminating the need for any prior atlases, lung segmentation or any subsequent user intervention. We evaluated our model using 84 chest CT scans from the LIDC and 154 pathological cases from the LTRC datasets. Our model achieved a Dice score of $0.939 \pm 0.02$ for the LIDC test set and $0.950 \pm 0.01$ for the LTRC test set, significantly outperforming a 2D U-net model and a 3D dense V-net. We further evaluated our model against 55 cases from the LOLA11 challenge, obtaining an average Dice score of 0.935—a performance level competitive to the best performing team with an average score of 0.938. Our extensive robustness analyses also demonstrate that our model can reliably segment both healthy and pathological lung lobes in CT scans from different vendors, and that our model is robust against configurations of CT scan reconstruction.
Tasks
Published	2019-02-18
URL	http://arxiv.org/abs/1902.06362v1
PDF	http://arxiv.org/pdf/1902.06362v1.pdf
PWC	https://paperswithcode.com/paper/automatic-segmentation-of-pulmonary-lobes
Repo
Framework

Sarcasm Detection using Hybrid Neural Network


Title	Sarcasm Detection using Hybrid Neural Network
Authors	Rishabh Misra, Prahal Arora
Abstract	Sarcasm Detection has enjoyed great interest from the research community, however the task of predicting sarcasm in a text remains an elusive problem for machines. Past studies mostly make use of twitter datasets collected using hashtag based supervision but such datasets are noisy in terms of labels and language. To overcome these shortcoming, we introduce a new dataset which contains news headlines from a sarcastic news website and a real news website. Next, we propose a hybrid Neural Network architecture with attention mechanism which provides insights about what actually makes sentences sarcastic. Through experiments, we show that the proposed model improves upon the baseline by ~ 5% in terms of classification accuracy.
Tasks	Sarcasm Detection
Published	2019-08-20
URL	https://arxiv.org/abs/1908.07414v1
PDF	https://arxiv.org/pdf/1908.07414v1.pdf
PWC	https://paperswithcode.com/paper/sarcasm-detection-using-hybrid-neural-network
Repo
Framework

Recursion, Probability, Convolution and Classification for Computations


Title	Recursion, Probability, Convolution and Classification for Computations
Authors	Mircea Namolaru, Thierry Goubier
Abstract	The main motivation of this work was practical, to offer computationally and theoretical scalable ways to structuring large classes of computation. It started from attempts to optimize R code for machine learning/artificial intelligence algorithms for huge data sets, that due to their size, should be handled into an incremental (online) fashion. Our target are large classes of relational (attribute based), mathematical (index based) or graph computations. We wanted to use powerful computation representations that emerged in AI (artificial intelligence)/ML (machine learning) as BN (Bayesian networks) and CNN (convolution neural networks). For the classes of computation addressed by us, and for our HPC (high performance computing) needs, the current solutions for translating computations into such representation need to be extended. Our results show that the classes of computation targeted by us, could be tree-structured, and a probability distribution (defining a DBN, i.e. Dynamic Bayesian Network) associated with it. More ever, this DBN may be viewed as a recursive CNN (Convolution Neural Network). Within this tree-like structure, classification in classes with size bounded (by a parameterizable may be performed. These results are at the core of very powerful, yet highly practically algorithms for restructuring and parallelizing the computations. The mathematical background required for an in depth presentation and exposing the full generality of our approach) is the subject of a subsequent paper. In this paper, we work in an limited (but important) framework that could be understood with rudiments of linear algebra and graph theory. The focus is in applicability, most of this paper discuss the usefulness of our approach for solving hard compilation problems related to automatic parallelism.
Tasks
Published	2019-07-22
URL	https://arxiv.org/abs/1908.04265v1
PDF	https://arxiv.org/pdf/1908.04265v1.pdf
PWC	https://paperswithcode.com/paper/recursion-probability-convolution-and
Repo
Framework

Semantic Correlation Promoted Shape-Variant Context for Segmentation


Title	Semantic Correlation Promoted Shape-Variant Context for Segmentation
Authors	Henghui Ding, Xudong Jiang, Bing Shuai, Ai Qun Liu, Gang Wang
Abstract	Context is essential for semantic segmentation. Due to the diverse shapes of objects and their complex layout in various scene images, the spatial scales and shapes of contexts for different objects have very large variation. It is thus ineffective or inefficient to aggregate various context information from a predefined fixed region. In this work, we propose to generate a scale- and shape-variant semantic mask for each pixel to confine its contextual region. To this end, we first propose a novel paired convolution to infer the semantic correlation of the pair and based on that to generate a shape mask. Using the inferred spatial scope of the contextual region, we propose a shape-variant convolution, of which the receptive field is controlled by the shape mask that varies with the appearance of input. In this way, the proposed network aggregates the context information of a pixel from its semantic-correlated region instead of a predefined fixed region. Furthermore, this work also proposes a labeling denoising model to reduce wrong predictions caused by the noisy low-level features. Without bells and whistles, the proposed segmentation network achieves new state-of-the-arts consistently on the six public segmentation datasets.
Tasks	Denoising, Semantic Segmentation
Published	2019-09-05
URL	https://arxiv.org/abs/1909.02651v1
PDF	https://arxiv.org/pdf/1909.02651v1.pdf
PWC	https://paperswithcode.com/paper/semantic-correlation-promoted-shape-variant-1
Repo
Framework

An Adaptive Similarity Measure to Tune Trust Influence in Memory-Based Collaborative Filtering


Title	An Adaptive Similarity Measure to Tune Trust Influence in Memory-Based Collaborative Filtering
Authors	Mohammad Reza Zarei, Mohammad R. Moosavi
Abstract	The aim of the recommender systems is to provide relevant and potentially interesting information to each user. This is fulfilled by utilizing the already recorded tendencies of similar users or detecting items similar to interested items of the user. Challenges such as data sparsity and cold start problem are addressed in recent studies. Utilizing social information not only enhances the prediction accuracy but also tackles the data sparseness challenges. In this paper, we investigate the impact of using direct and indirect trust information in a memory-based collaborative filtering recommender system. An adaptive similarity measure is proposed and the contribution of social information is tuned using two learning schemes, greedy and gradient-based optimization. The results of the proposed method are compared with state-of-the-art memory-based and model-based CF approaches on two real-world datasets, Epinions and FilmTrust. The experiments show that our method is quite effective in designing an accurate and comprehensive recommender system.
Tasks	Recommendation Systems
Published	2019-12-18
URL	https://arxiv.org/abs/1912.08934v1
PDF	https://arxiv.org/pdf/1912.08934v1.pdf
PWC	https://paperswithcode.com/paper/an-adaptive-similarity-measure-to-tune-trust
Repo
Framework

Predicting drug-target interaction using 3D structure-embedded graph representations from graph neural networks


Title	Predicting drug-target interaction using 3D structure-embedded graph representations from graph neural networks
Authors	Jaechang Lim, Seongok Ryu, Kyubyong Park, Yo Joong Choe, Jiyeon Ham, Woo Youn Kim
Abstract	Accurate prediction of drug-target interaction (DTI) is essential for in silico drug design. For the purpose, we propose a novel approach for predicting DTI using a GNN that directly incorporates the 3D structure of a protein-ligand complex. We also apply a distance-aware graph attention algorithm with gate augmentation to increase the performance of our model. As a result, our model shows better performance than docking and other deep learning methods for both virtual screening and pose prediction. In addition, our model can reproduce the natural population distribution of active molecules and inactive molecules.
Tasks	Pose Prediction
Published	2019-04-17
URL	http://arxiv.org/abs/1904.08144v1
PDF	http://arxiv.org/pdf/1904.08144v1.pdf
PWC	https://paperswithcode.com/paper/predicting-drug-target-interaction-using-3d
Repo
Framework