Paper Group NANR 76
Node Embeddings for Graph Merging: Case of Knowledge Graph Construction. Learning a Mixture of Granularity-Specific Experts for Fine-Grained Categorization. Learning Local Descriptors With a CDF-Based Dynamic Soft Margin. Huawei’s NMT Systems for the WMT 2019 Biomedical Translation Task. Understanding & Generalizing AlphaGo Zero. Multiclass Queuein …
Node Embeddings for Graph Merging: Case of Knowledge Graph Construction
| Title | Node Embeddings for Graph Merging: Case of Knowledge Graph Construction |
| Authors | Ida Szubert, Mark Steedman |
| Abstract | Combining two graphs requires merging the nodes which are counterparts of each other. In this process errors occur, resulting in incorrect merging or incorrect failure to merge. We find a high prevalence of such errors when using AskNET, an algorithm for building Knowledge Graphs from text corpora. AskNET node matching method uses string similarity, which we propose to replace with vector embedding similarity. We explore graph-based and word-based embedding models and show an overall error reduction of from 56{%} to 23.6{%}, with a reduction of over a half in both types of incorrect node matching. |
| Tasks | graph construction, Knowledge Graphs |
| Published | 2019-11-01 |
| URL | https://www.aclweb.org/anthology/D19-5321/ |
| https://www.aclweb.org/anthology/D19-5321 | |
| PWC | https://paperswithcode.com/paper/node-embeddings-for-graph-merging-case-of |
| Repo | |
| Framework | |
Learning a Mixture of Granularity-Specific Experts for Fine-Grained Categorization
| Title | Learning a Mixture of Granularity-Specific Experts for Fine-Grained Categorization |
| Authors | Lianbo Zhang, Shaoli Huang, Wei Liu, Dacheng Tao |
| Abstract | We aim to divide the problem space of fine-grained recognition into some specific regions. To achieve this, we develop a unified framework based on a mixture of experts. Due to limited data available for the fine-grained recognition problem, it is not feasible to learn diverse experts by using a data division strategy. To tackle the problem, we promote diversity among experts by combing an expert gradually-enhanced learning strategy and a Kullback-Leibler divergence based constraint. The strategy learns new experts on the dataset with the prior knowledge from former experts and adds them to the model sequentially, while the introduced constraint forces the experts to produce diverse prediction distribution. These drive the experts to learn the task from different aspects, making them specialized in different subspace problems. Experiments show that the resulting model improves the classification performance and achieves the state-of-the-art performance on several fine-grained benchmark datasets. |
| Tasks | Fine-Grained Image Classification |
| Published | 2019-10-01 |
| URL | http://openaccess.thecvf.com/content_ICCV_2019/html/Zhang_Learning_a_Mixture_of_Granularity-Specific_Experts_for_Fine-Grained_Categorization_ICCV_2019_paper.html |
| http://openaccess.thecvf.com/content_ICCV_2019/papers/Zhang_Learning_a_Mixture_of_Granularity-Specific_Experts_for_Fine-Grained_Categorization_ICCV_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/learning-a-mixture-of-granularity-specific |
| Repo | |
| Framework | |
Learning Local Descriptors With a CDF-Based Dynamic Soft Margin
| Title | Learning Local Descriptors With a CDF-Based Dynamic Soft Margin |
| Authors | Linguang Zhang, Szymon Rusinkiewicz |
| Abstract | The triplet loss is adopted by a variety of learning tasks, such as local feature descriptor learning. However, its standard formulation with a hard margin only leverages part of the training data in each mini-batch. Moreover, the margin is often empirically chosen or determined through computationally expensive validation, and stays unchanged during the entire training session. In this work, we propose a simple yet effective method to overcome the above limitations. The core idea is to replace the hard margin with a non-parametric soft margin, which is dynamically updated. The major observation is that the difficulty of a triplet can be inferred from the cumulative distribution function of the triplets’ signed distances to the decision boundary. We demonstrate through experiments on both real-valued and binary local feature descriptors that our method leads to state-of-the-art performance on popular benchmarks, while eliminating the need to determine the best margin. |
| Tasks | |
| Published | 2019-10-01 |
| URL | http://openaccess.thecvf.com/content_ICCV_2019/html/Zhang_Learning_Local_Descriptors_With_a_CDF-Based_Dynamic_Soft_Margin_ICCV_2019_paper.html |
| http://openaccess.thecvf.com/content_ICCV_2019/papers/Zhang_Learning_Local_Descriptors_With_a_CDF-Based_Dynamic_Soft_Margin_ICCV_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/learning-local-descriptors-with-a-cdf-based |
| Repo | |
| Framework | |
Huawei’s NMT Systems for the WMT 2019 Biomedical Translation Task
| Title | Huawei’s NMT Systems for the WMT 2019 Biomedical Translation Task |
| Authors | Wei Peng, Jianfeng Liu, Liangyou Li, Qun Liu |
| Abstract | This paper describes Huawei{'}s neural machine translation systems for the WMT 2019 biomedical translation shared task. We trained and fine-tuned our systems on a combination of out-of-domain and in-domain parallel corpora for six translation directions covering English{–}Chinese, English{–}French and English{–}German language pairs. Our submitted systems achieve the best BLEU scores on English{–}French and English{–}German language pairs according to the official evaluation results. In the English{–}Chinese translation task, our systems are in the second place. The enhanced performance is attributed to more in-domain training and more sophisticated models developed. Development of translation models and transfer learning (or domain adaptation) methods has significantly contributed to the progress of the task. |
| Tasks | Domain Adaptation, Machine Translation, Transfer Learning |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-5420/ |
| https://www.aclweb.org/anthology/W19-5420 | |
| PWC | https://paperswithcode.com/paper/huaweis-nmt-systems-for-the-wmt-2019 |
| Repo | |
| Framework | |
Understanding & Generalizing AlphaGo Zero
| Title | Understanding & Generalizing AlphaGo Zero |
| Authors | Ravichandra Addanki, Mohammad Alizadeh, Shaileshh Bojja Venkatakrishnan, Devavrat Shah, Qiaomin Xie, Zhi Xu |
| Abstract | AlphaGo Zero (AGZ) introduced a new {\em tabula rasa} reinforcement learning algorithm that has achieved superhuman performance in the games of Go, Chess, and Shogi with no prior knowledge other than the rules of the game. This success naturally begs the question whether it is possible to develop similar high-performance reinforcement learning algorithms for generic sequential decision-making problems (beyond two-player games), using only the constraints of the environment as the ``rules.’’ To address this challenge, we start by taking steps towards developing a formal understanding of AGZ. AGZ includes two key innovations: (1) it learns a policy (represented as a neural network) using {\em supervised learning} with cross-entropy loss from samples generated via Monte-Carlo Tree Search (MCTS); (2) it uses {\em self-play} to learn without training data. We argue that the self-play in AGZ corresponds to learning a Nash equilibrium for the two-player game; and the supervised learning with MCTS is attempting to learn the policy corresponding to the Nash equilibrium, by establishing a novel bound on the difference between the expected return achieved by two policies in terms of the expected KL divergence (cross-entropy) of their induced distributions. To extend AGZ to generic sequential decision-making problems, we introduce a {\em robust MDP} framework, in which the agent and nature effectively play a zero-sum game: the agent aims to take actions to maximize reward while nature seeks state transitions, subject to the constraints of that environment, that minimize the agent’s reward. For a challenging network scheduling domain, we find that AGZ within the robust MDP framework provides near-optimal performance, matching one of the best known scheduling policies that has taken the networking community three decades of intensive research to develop. | |
| Tasks | Decision Making |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=rkxtl3C5YX |
| https://openreview.net/pdf?id=rkxtl3C5YX | |
| PWC | https://paperswithcode.com/paper/understanding-generalizing-alphago-zero |
| Repo | |
| Framework | |
Multiclass Queueing Network Modeling and Traffic Flow Analysis for SDN-enabled Mobile Core Networks with Network Slicing
| Title | Multiclass Queueing Network Modeling and Traffic Flow Analysis for SDN-enabled Mobile Core Networks with Network Slicing |
| Authors | Santhosha Kamath, Sanjay Singh, M Sathish Kumar |
| Abstract | The back-haul networks of 5G are formed by heterogeneous links which need to handle massive traffic. The service providers are not able to provide good QoS for their users. The technology like Software Defined Networks(SDN) and Network Slicing helps a little for a service provider to providing QoS for multiple links. The service providers face a challenge in the efficient utilization of resources to fulfill the QoS requirement of users to comply with the growth and thereby increasing the revenue. These problems require an accurate traffic model to determine the steady-state of the system. The proposed model uses an architecture that has the combination of two technologies: SDN and network slicing, which empowers an administrator a flexible, programmable network, and the best management of network resources. Heterogeneous application is well managed by creating multiple logical networks called slicing. The slicing can be modeled using multi-class queuing networks. These technologies encourage service providers to fulfill QoS and revenue growth. To leverage the benefits of these technologies in allocating QoS is to identify the performance of the system, which requires a precise model of traffic to decide the steady-state condition of the framework. In this paper, we focus on SDN and slicing in mobile networks and quantify the performance measure considering an in-band OpenFlow architecture for a single node and homogeneous traffic class, which is further extended to the multi-class heterogeneous class queuing model and analyzed. The results obtained help a service provider to monitor the utilization of resources in every node by every class of core network, which in turn helps to allocate the resources precisely to fulfill QoS requirements. |
| Tasks | |
| Published | 2019-12-13 |
| URL | https://ieeexplore.ieee.org/abstract/document/8932489 |
| https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8932489 | |
| PWC | https://paperswithcode.com/paper/multiclass-queueing-network-modeling-and |
| Repo | |
| Framework | |
The Impact of Rule-Based Text Generation on the Quality of Abstractive Summaries
| Title | The Impact of Rule-Based Text Generation on the Quality of Abstractive Summaries |
| Authors | Tatiana Vodolazova, Elena Lloret |
| Abstract | In this paper we describe how an abstractive text summarization method improved the informativeness of automatic summaries by integrating syntactic text simplification, subject-verb-object concept frequency scoring and a set of rules that transform text into its semantic representation. We analyzed the impact of each component of our approach on the quality of generated summaries and tested it on DUC 2002 dataset. Our experiments showed that our approach outperformed other state-of-the-art abstractive methods while maintaining acceptable linguistic quality and redundancy rate. |
| Tasks | Abstractive Text Summarization, Text Generation, Text Simplification, Text Summarization |
| Published | 2019-09-01 |
| URL | https://www.aclweb.org/anthology/R19-1146/ |
| https://www.aclweb.org/anthology/R19-1146 | |
| PWC | https://paperswithcode.com/paper/the-impact-of-rule-based-text-generation-on |
| Repo | |
| Framework | |
Catastrophic Child’s Play: Easy to Perform, Hard to Defend Adversarial Attacks
| Title | Catastrophic Child’s Play: Easy to Perform, Hard to Defend Adversarial Attacks |
| Authors | Chih-Hui Ho, Brandon Leung, Erik Sandstrom, Yen Chang, Nuno Vasconcelos |
| Abstract | The problem of adversarial CNN attacks is considered, with an emphasis on attacks that are trivial to perform but difficult to defend. A framework for the study of such attacks is proposed, using real world object manipulations. Unlike most works in the past, this framework supports the design of attacks based on both small and large image perturbations, implemented by camera shake and pose variation. A setup is proposed for the collection of such perturbations and determination of their perceptibility. It is argued that perceptibility depends on context, and a distinction is made between imperceptible and semantically imperceptible perturbations. While the former survives image comparisons, the latter are perceptible but have no impact on human object recognition. A procedure is proposed to determine the perceptibility of perturbations using Turk experiments, and a dataset of both perturbation classes which enables replicable studies of object manipulation attacks, is assembled. Experiments using defenses based on many datasets, CNN models, and algorithms from the literature elucidate the difficulty of defending these attacks – in fact, none of the existing defenses is found effective against them. Better results are achieved with real world data augmentation, but even this is not foolproof. These results confirm the hypothesis that current CNNs are vulnerable to attacks implementable even by a child, and that such attacks may prove difficult to defend. |
| Tasks | Data Augmentation, Object Recognition |
| Published | 2019-06-01 |
| URL | http://openaccess.thecvf.com/content_CVPR_2019/html/Ho_Catastrophic_Childs_Play_Easy_to_Perform_Hard_to_Defend_Adversarial_CVPR_2019_paper.html |
| http://openaccess.thecvf.com/content_CVPR_2019/papers/Ho_Catastrophic_Childs_Play_Easy_to_Perform_Hard_to_Defend_Adversarial_CVPR_2019_paper.pdf | |
| PWC | https://paperswithcode.com/paper/catastrophic-childs-play-easy-to-perform-hard |
| Repo | |
| Framework | |
Content Customization for Micro Learning using Human Augmented AI Techniques
| Title | Content Customization for Micro Learning using Human Augmented AI Techniques |
| Authors | Ayush Shah, Tamer Abuelsaad, Jae-Wook Ahn, Prasenjit Dey, Ravi Kokku, Ruhi Sharma Mittal, Aditya Vempaty, Mourvi Sharma |
| Abstract | Visual content has been proven to be effective for micro-learning compared to other media. In this paper, we discuss leveraging this observation in our efforts to build audio-visual content for young learners{'} vocabulary learning. We attempt to tackle two major issues in the process of traditional visual curation tasks. Generic learning videos do not necessarily satisfy the unique context of a learner and/or an educator, and hence may not result in maximal learning outcomes. Also, manual video curation by educators is a highly labor-intensive process. To this end, we present a customizable micro-learning audio-visual content curation tool that is designed to reduce the human (educator) effort in creating just-in-time learning videos from a textual description (learning script). This provides educators with control of the content while preparing the learning scripts, and in turn can also be customized to capture the desired learning objectives and outcomes. As a use case, we automatically generate learning videos with British National Corpus{'} (BNC) frequently spoken vocabulary words and evaluate them with experts. They positively recommended the generated learning videos with an average rating of 4.25 on a Likert scale of 5 points. The inter-annotator agreement between the experts for the video quality was substantial (Fleiss Kappa=0.62) with an overall agreement of 81{%}. |
| Tasks | |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-4434/ |
| https://www.aclweb.org/anthology/W19-4434 | |
| PWC | https://paperswithcode.com/paper/content-customization-for-micro-learning |
| Repo | |
| Framework | |
Modeling language learning using specialized Elo rating
| Title | Modeling language learning using specialized Elo rating |
| Authors | Jue Hou, Koppatz Maximilian, Jos{'e} Mar{'\i}a Hoya Quecedo, Nataliya Stoyanova, Roman Yangarber |
| Abstract | Automatic assessment of the proficiency levels of the learner is a critical part of Intelligent Tutoring Systems. We present methods for assessment in the context of language learning. We use a specialized Elo formula used in conjunction with educational data mining. We simultaneously obtain ratings for the proficiency of the learners and for the difficulty of the linguistic concepts that the learners are trying to master. From the same data we also learn a graph structure representing a domain model capturing the relations among the concepts. This application of Elo provides ratings for learners and concepts which correlate well with subjective proficiency levels of the learners and difficulty levels of the concepts. |
| Tasks | |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-4451/ |
| https://www.aclweb.org/anthology/W19-4451 | |
| PWC | https://paperswithcode.com/paper/modeling-language-learning-using-specialized |
| Repo | |
| Framework | |
Large-scale Machine Translation Evaluation of the iADAATPA Project
| Title | Large-scale Machine Translation Evaluation of the iADAATPA Project |
| Authors | Sheila Castilho, Nat{'a}lia Resende, Federico Gaspari, Andy Way, Tony O{'}Dowd, Marek Mazur, Manuel Herranz, Alex Helle, Gema Ram{'\i}rez-S{'a}nchez, V{'\i}ctor S{'a}nchez-Cartagena, M{=a}rcis Pinnis, Valters {\v{S}}ics |
| Abstract | |
| Tasks | Machine Translation |
| Published | 2019-08-01 |
| URL | https://www.aclweb.org/anthology/W19-6732/ |
| https://www.aclweb.org/anthology/W19-6732 | |
| PWC | https://paperswithcode.com/paper/large-scale-machine-translation-evaluation-of |
| Repo | |
| Framework | |
Sample-efficient policy learning in multi-agent Reinforcement Learning via meta-learning
| Title | Sample-efficient policy learning in multi-agent Reinforcement Learning via meta-learning |
| Authors | Jialian Li, Hang Su, Jun Zhu |
| Abstract | To gain high rewards in muti-agent scenes, it is sometimes necessary to understand other agents and make corresponding optimal decisions. We can solve these tasks by first building models for other agents and then finding the optimal policy with these models. To get an accurate model, many observations are needed and this can be sample-inefficient. What’s more, the learned model and policy can overfit to current agents and cannot generalize if the other agents are replaced by new agents. In many practical situations, each agent we face can be considered as a sample from a population with a fixed but unknown distribution. Thus we can treat the task against some specific agents as a task sampled from a task distribution. We apply meta-learning method to build models and learn policies. Therefore when new agents come, we can adapt to them efficiently. Experiments on grid games show that our method can quickly get high rewards. |
| Tasks | Meta-Learning, Multi-agent Reinforcement Learning |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=r1fiFs09YX |
| https://openreview.net/pdf?id=r1fiFs09YX | |
| PWC | https://paperswithcode.com/paper/sample-efficient-policy-learning-in-multi |
| Repo | |
| Framework | |
Visual Sequence Learning in Hierarchical Prediction Networks and Primate Visual Cortex
| Title | Visual Sequence Learning in Hierarchical Prediction Networks and Primate Visual Cortex |
| Authors | Jielin Qiu, Ge Huang, Tai Sing Lee |
| Abstract | In this paper we developed a computational hierarchical network model to understand the spatiotemporal sequence learning effects observed in the primate visual cortex. The model is a hierarchical recurrent neural model that learns to predict video sequences using the incoming video signals as teaching signals. The model performs fast feedforward analysis using a deep convolutional neural network with sparse convolution and feedback synthesis using a stack of LSTM modules. The network learns a representational hierarchy by minimizing its prediction errors of the incoming signals at each level of the hierarchy. We found that recurrent feedback in this network lead to the development of semantic cluster of global movement patterns in the population codes of the units at the lower levels of the hierarchy. These representations facilitate the learning of relationship among movement patterns, yielding state-of-the-art performance in long range video sequence predictions on benchmark datasets. Without further tuning, this model automatically exhibits the neurophysiological correlates of visual sequence memories that we observed in the early visual cortex of awake monkeys, suggesting the principle of self-supervised prediction learning might be relevant to understanding the cortical mechanisms of representational learning. |
| Tasks | |
| Published | 2019-12-01 |
| URL | http://papers.nips.cc/paper/8534-visual-sequence-learning-in-hierarchical-prediction-networks-and-primate-visual-cortex |
| http://papers.nips.cc/paper/8534-visual-sequence-learning-in-hierarchical-prediction-networks-and-primate-visual-cortex.pdf | |
| PWC | https://paperswithcode.com/paper/visual-sequence-learning-in-hierarchical |
| Repo | |
| Framework | |
Optimal Control Via Neural Networks: A Convex Approach
| Title | Optimal Control Via Neural Networks: A Convex Approach |
| Authors | Yize Chen, Yuanyuan Shi, Baosen Zhang |
| Abstract | Control of complex systems involves both system identification and controller design. Deep neural networks have proven to be successful in many identification tasks, however, from model-based control perspective, these networks are difficult to work with because they are typically nonlinear and nonconvex. Therefore many systems are still identified and controlled based on simple linear models despite their poor representation capability. In this paper we bridge the gap between model accuracy and control tractability faced by neural networks, by explicitly constructing networks that are convex with respect to their inputs. We show that these input convex networks can be trained to obtain accurate models of complex physical systems. In particular, we design input convex recurrent neural networks to capture temporal behavior of dynamical systems. Then optimal controllers can be achieved via solving a convex model predictive control problem. Experiment results demonstrate the good potential of the proposed input convex neural network based approach in a variety of control applications. In particular we show that in the MuJoCo locomotion tasks, we could achieve over 10% higher performance using 5 times less time compared with state-of-the-art model-based reinforcement learning method; and in the building HVAC control example, our method achieved up to 20% energy reduction compared with classic linear models. |
| Tasks | |
| Published | 2019-05-01 |
| URL | https://openreview.net/forum?id=H1MW72AcK7 |
| https://openreview.net/pdf?id=H1MW72AcK7 | |
| PWC | https://paperswithcode.com/paper/optimal-control-via-neural-networks-a-convex |
| Repo | |
| Framework | |
Augmentic Compositional Models for Knowledge Base Completion Using Gradient Representations
| Title | Augmentic Compositional Models for Knowledge Base Completion Using Gradient Representations |
| Authors | Matthias R. Lalisse, Paul Smolensky |
| Abstract | |
| Tasks | Knowledge Base Completion |
| Published | 2019-01-01 |
| URL | https://www.aclweb.org/anthology/W19-0126/ |
| https://www.aclweb.org/anthology/W19-0126 | |
| PWC | https://paperswithcode.com/paper/augmentic-compositional-models-for-knowledge |
| Repo | |
| Framework | |
