检索结果-内蒙古大学图书馆

Neural graph matching network: Learning Lawler's quadratic assignment problem with extension to hypergraph and multiple-graph matching

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Wang, Runzhong Yan, Junchi Yang, Xiaokang Department of Computer Science and Engineering MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai200240 China

Graph matching involves combinatorial optimization based on edge-to-edge affinity matrix, which can be generally formulated as Lawler's Quadratic Assignment Problem (QAP). This paper presents a QAP network directly learning with the affinity matrix (equivalently the association graph) whereby the matching problem is translated into a constrained vertex classification task. The association graph is learned by an embedding network for vertex classification, followed by Sinkhorn normalization and a cross-entropy loss for end-to-end learning. We further improve the embedding model on association graph by introducing Sinkhorn based matching-aware constraint, as well as dummy nodes to deal with unequal sizes of graphs. To our best knowledge, this is one of the first network to directly learn with the general Lawler's QAP. In contrast, recent deep matching methods focus on the learning of node/edge features in two graphs respectively. We also show how to extend our network to hypergraph matching, and matching of multiple graphs. Experimental results on both synthetic graphs and real-world images show its effectiveness. For pure QAP tasks on synthetic data and QAPLIB benchmark, our method can perform competitively and even surpass state-of-the-art graph matching and QAP solvers with notable less time cost. We provide a project homepage at http://***/project/NGM/***. Copyright © 2019, The Authors. All rights reserved.

关键词： Combinatorial optimization

Retrospective Reader for Machine Reading Comprehension

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhang, Zhuosheng Yang, Junjie Zhao, Hai Department of Computer Science and Engineering Shanghai Jiao Tong University Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University Shanghai China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China SJTU-ParisTech Elite Institute of Technology Shanghai Jiao Tong University Shanghai China

Machine reading comprehension (MRC) is an ai challenge that requires machine to determine the correct answers to questions based on a given passage. MRC systems must not only answer question when necessary but also distinguish when no answer is available according to the given passage and then tactfully abstain from answering. When unanswerable questions are involved in the MRC task, an essential verification module called verifier is especially required in addition to the encoder, though the latest practice on MRC modeling still most benefits from adopting well pre-trained language models as the encoder block by only focusing on the "reading". This paper devotes itself to exploring better verifier design for the MRC task with unanswerable questions. Inspired by how humans solve reading comprehension questions, we proposed a retrospective reader (Retro-Reader) that integrates two stages of reading and verification strategies: 1) sketchy reading that briefly investigates the overall interactions of passage and question, and yield an initial judgment;2) intensive reading that verifies the answer and gives the final prediction. The proposed reader is evaluated on two benchmark MRC challenge datasets SQuAD2.0 and NewsQA, achieving new state-of-the-art results. Significance tests show that our model is significantly better than the strong ALBERT baseline. A series of analysis is also conducted to interpret the effectiveness of the proposed reader. Copyright © 2020, The Authors. All rights reserved.

关键词： Signal encoding

Learning combinatorial embedding networks for deep graph matching

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Wang, Runzhong Yan, Junchi Yang, Xiaokang Department of Computer Science and Engineering Shanghai Jiao Tong University MoE Key Lab of Artificial Intelligence Ai Institute Shanghai Jiao Tong University

Graph matching refers to finding node correspondence between graphs, such that the corresponding node and edge's affinity can be maximized. In addition with its NP-completeness nature, another important challenge is effective modeling of the node-wise and structure-wise affinity across graphs and the resulting objective, to guide the matching procedure effectively finding the true matching against noises. To this end, this paper devises an end-to-end differentiable deep network pipeline to learn the affinity for graph matching. It involves a supervised permutation loss regarding with node correspondence to capture the combinatorial nature for graph matching. Meanwhile deep graph embedding models are adopted to parameterize both intra-graph and cross-graph affinity functions, instead of the traditional shallow and simple parametric forms e.g. a Gaussian kernel. The embedding can also effectively capture the higher-order structure beyond second-order edges. The permutation loss model is agnostic to the number of nodes, and the embedding model is shared among nodes such that the network allows for varying numbers of nodes in graphs for training and inference. Moreover, our network is class-agnostic with some generalization capability across different categories. All these features are welcomed for real-world applications. Experiments show its superiority against state-of-the-art graph matching learning methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep learning

Learning Combinatorial Embedding Networks for Deep Graph Matching

学校读者我要写书评

暂无评论

Learning Combinatorial Embedding Networks for Deep Graph Mat...

International Conference on computer Vision (ICCV)

作者： Runzhong Wang Junchi Yan Xiaokang Yang Department of Computer Science and Engineering Shanghai Jiao Tong University MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University

ISBN: (数字)9781728148038

ISBN: (纸本)9781728148045

关键词： Peer-to-peer computing Feature extraction Tensile stress Neural networks Kernel Training Pipelines

Explicit Shape Encoding for Real-Time Instance Segmentation

学校读者我要写书评

暂无评论

Explicit Shape Encoding for Real-Time Instance Segmentation

International Conference on computer Vision (ICCV)

作者： Wenqiang Xu Haiyang Wang Fubo Qi Cewu Lu Department of Computer Science and Engineering Shanghai Jiao Tong University member of MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University and SJTU-SenseTime AI lab.

ISBN: (数字)9781728148038

ISBN: (纸本)9781728148045

In this paper, we propose a novel top-down instance segmentation framework based on explicit shape encoding, named \textbf{ESE-Seg}. It largely reduces the computational consumption of the instance segmentation by explicitly decoding the multiple object shapes with tensor operations, thus performs the instance segmentation at almost the same speed as the object detection. ESE-Seg is based on a novel shape signature Inner-center Radius (IR), Chebyshev polynomial fitting and the strong modern object detectors. ESE-Seg with YOLOv3 outperforms the Mask R-CNN on Pascal VOC 2012 at mAP r @0.5 while 7 times faster.

关键词： Shape Object detection Decoding Encoding Chebyshev approximation Detectors Training

Impact of indirect transitions on valley polarization in WS2 and WSe2

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Godiksen, Rasmus H. Wang, Shaojun Raziman, T.V. Rivas, Jaime Gómez Curto, Alberto G. Dep. Applied Physics Institute for Photonic Integration Eindhoven University of Technology Eindhoven Netherlands MOE Key Lab. of Modern Optical Technologies Jiangsu Key Lab. of Advanced Optical Manufacturing Technologies School of Optoelectronic Science and Engineering Soochow University Suzhou215006 China Photonics Research Group Ghent University-imec Ghent Belgium Center for Nano- and Biophotonics Ghent University Ghent Belgium

Controlling the momentum of carriers in semiconductors, known as valley polarization, is a new resource for optoelectronics and information technologies. Materials exhibiting high polarization are needed for valley-based devices. Few-layer WS2 shows a remarkable spin-valley polarization above 90%, even at room temperature. In stark contrast, polarization is absent for few-layer WSe2 despite the expected material similarities. Here, we explain the origin of valley polarization in both materials based on the interplay between two indirect optical transitions. We show that the relative energy minima at the Λ- and K-valleys in the conduction band determine the spin-valley polarization of the direct K-K transition. Polarization appears as the energy of the K-valley rises above the Λ-valley as a function of temperature and number of layers. Our results advance the understanding of the high spin-valley polarization in WS2. This insight will impact the design of both passive and tunable valleytronic devices operating at room temperature. Copyright © 2020, The Authors. All rights reserved.

关键词： Polarization

Concurrent parsing of constituency and dep.ndency

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Zhou, Junru Zhang, Shuailiang Zhao, Hai Department of Computer Science and Engineering Key Lab of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China

Constituent and dep.ndency representation for syntactic structure share a lot of linguistic and computational characteristics, this paper thus makes the first attempt by introducing a new model that is capable of parsing constituent and dep.ndency at the same time, so that lets either of the parsers enhance each other. Especially, we evaluate the effect of different shared network components and empirically verify that dep.ndency parsing may be much more beneficial from constituent parsing structure. The proposed parser achieves new stateof-the-art performance for both parsing tasks, constituent and dep.ndency on PTB and CTB benchmarks. Copyright © 2019, The Authors. All rights reserved.

关键词： Syntactics

Multi-span style extraction for generative reading comprehension

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Yang, Junjie Zhang, Zhuosheng Zhao, Hai SJTU-ParisTech Elite Institute of Technology Shanghai Jiao Tong University Shanghai China Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction Cognitive Engineering Shanghai Jiao Tong University Shanghai China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China

Generative machine reading comprehension (MRC) requires a model to generate well-formed answers. For this type of MRC, answer generation method is crucial to the model performance. However, generative models, which are supposed to be the right model for the task, in generally perform poorly. At the same time, single-span extraction models have been proven effective for extractive MRC, where the answer is constrained to a single span in the passage. Nevertheless, they generally suffer from generating incomplete answers or introducing redundant words when applied to the generative MRC. Thus, we extend the single-span extraction method to multi-span, proposing a new framework which enables generative MRC to be smoothly solved as multi-span extraction. Thorough experiments demonstrate that this novel approach can alleviate the dilemma between generative models and single-span models and produce answers with better-formed syntax and semantics. We will open-source our code for the research community. Copyright © 2020, The Authors. All rights reserved.

关键词： Extraction

Margin matters: Towards more discriminative deep neural network embeddings for speaker recognition

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Xiang, Xu Wang, Shuai Huang, Houjun Qian, Yanmin Yu, Kai MoE Key Lab of Artificial Intelligence SpeechLab Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai AI Speech Co. Ltd

Recently, speaker embeddings extracted from a speaker discriminative deep neural network (DNN) yield better performance than the conventional methods such as i-vector. In most cases, the DNN speaker classifier is trained using cross entropy loss with softmax. However, this kind of loss function does not explicitly encourage inter-class separability and intraclass compactness. As a result, the embeddings are not optimal for speaker recognition tasks. In this paper, to address this issue, three different margin based losses which not only separate classes but also demand a fixed margin between classes are introduced to deep speaker embedding learning. It could be demonstrated that the margin is the key to obtain more discriminative speaker embeddings. Experiments are conducted on two public text indep.ndent tasks: VoxCeleb1 and Speaker in The Wild (SITW). The proposed approach can achieve the state-ofthe- art performance, with 25% 30% equal error rate (EER) reduction on both tasks when compared to strong baselines using cross entropy loss with softmax, obtaining 2.238% EER on VoxCeleb1 test set and 2.761% EER on SITW core-core test set, respectively. Copyright © 2019, The Authors. All rights reserved.

关键词： Embeddings