Filters








204,314 Hits in 3.2 sec

Subspace Match Probably Does Not Accurately Assess the Similarity of Learned Representations [article]

Jeremiah Johnson
2019 arXiv   pre-print
To better understand this, subspace match was recently proposed as a method for assessing the similarity of the representations learned by neural networks.  ...  Learning informative representations of data is one of the primary goals of deep learning, but there is still little understanding as to what representations a neural network actually learns.  ...  Indeed, convolutional neural networks are explicitly designed to learn a hierarchical set of representations of images, and aside from only being used for image classification, these hierarchies of representations  ... 
arXiv:1901.00884v1 fatcat:uvqrcpxj5ne7jbokdgijrmihba

Neural Networks for Entity Matching: A Survey

Nils Barlaug, Jon Atle Gulla
2021 ACM Transactions on Knowledge Discovery from Data  
We also discuss contributions from deep learning in entity matching compared to traditional methods, and propose a taxonomy of deep neural networks for entity matching.  ...  Specifically, we identify which steps of the entity matching process existing work have targeted using neural networks, and provide an overview of the different techniques used at each step.  ...  ACKNOWLEDGMENTS This work is supported by Cognite and the Research Council of Norway under Project 298998. We thank the reviewers for valuable feedback and Carl F. Straumsheim for proofreading.  ... 
doi:10.1145/3442200 fatcat:odcz4lompjh3pghjzcpjhoeb2u

Convolutional Deep Neural Networks for Document-Based Question Answering [chapter]

Jian Fu, Xipeng Qiu, Xuanjing Huang
2016 Lecture Notes in Computer Science  
In this article, we present a convolutional neural network based architecture to learn feature representations of each questionanswer pair and compute its match score.  ...  Document-based Question Answering aims to compute the similarity or relevance between two texts: question and answer.  ...  By the way, attention mechanism and similarity match can be put into the top or bottom or both top and bottom layer of the neural network [18] .  ... 
doi:10.1007/978-3-319-50496-4_71 fatcat:e3k47u2oo5gancuwzdcyl3bzgq

Getting Started with Neural Models for Semantic Matching in Web Search [article]

Kezban Dilek Onal, Ismail Sengor Altingovde, Pinar Karagoz, Maarten de Rijke
2016 arXiv   pre-print
Such representations enable powerful semantic matching methods. This survey is meant as an introduction to the use of neural models for semantic matching.  ...  We detail the required background and terminology, a taxonomy grouping the rapidly growing body of work in the area, and then survey work on neural models for semantic matching in the context of three  ...  Learn to match In Table 6 , neural learn to match models for ad retrieval are summarized with the type of TTU pairs and the type of the neural network utilized as the SCN.  ... 
arXiv:1611.03305v1 fatcat:agdgj7allbczxcyteuomswn574

Neural Networks for Information Retrieval

Tom Kenter, Alexey Borisov, Christophe Van Gysel, Mostafa Dehghani, Maarten de Rijke, Bhaskar Mitra
2018 Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining - WSDM '18  
The aim of this full-day tutorial is to give a clear overview of current tried-and-trusted neural methods in IR and how they bene t IR research.  ...  Machine learning plays a role in many aspects of modern IR systems, and deep learning is applied in all of them.  ...  local representations -Use both local and global representations for query-document matching.  ... 
doi:10.1145/3159652.3162009 dblp:conf/wsdm/KenterBGDRM18 fatcat:ybdeuuxcbnh2np34k3y4ve5ovu

Neural Networks for Information Retrieval

Tom Kenter, Alexey Borisov, Christophe Van Gysel, Mostafa Dehghani, Maarten de Rijke, Bhaskar Mitra
2017 Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '17  
The aim of this full-day tutorial is to give a clear overview of current tried-and-trusted neural methods in IR and how they bene t IR research.  ...  Machine learning plays a role in many aspects of modern IR systems, and deep learning is applied in all of them.  ...  local representations -Use both local and global representations for query-document matching.  ... 
doi:10.1145/3077136.3082062 dblp:conf/sigir/KenterBGDRM17 fatcat:yxuiajzjlfaixlnhc6rrsud6ry

Neural Networks for Information Retrieval [article]

Tom Kenter, Alexey Borisov, Christophe Van Gysel, Mostafa Dehghani, Maarten de Rijke, Bhaskar Mitra
2017 arXiv   pre-print
The aim of this full-day tutorial is to give a clear overview of current tried-and-trusted neural methods in IR and how they benefit IR research.  ...  Machine learning plays a role in many aspects of modern IR systems, and deep learning is applied in all of them.  ...  local representations -Use both local and global representations for query-document matching.  ... 
arXiv:1707.04242v1 fatcat:4idscmq26fa5bjupldwuyghq4m

Improved object recognition using neural networks trained to mimic the brain's statistical properties [article]

Callie Federer and Haoyan Xu and Alona Fyshe and Joel Zylberberg
2020 arXiv   pre-print
The current state-of-the-art object recognition algorithms, deep convolutional neural networks (DCNNs), are inspired by the architecture of the mammalian visual system, and are capable of human-level performance  ...  To test this, we trained DCNNs on a composite task, wherein networks were trained to: a) classify images of objects; while b) having intermediate representations that resemble those observed in neural  ...  While the representations of images from different categories remain co-mingled at this low level of the neural network, the representations are more varied for the network trained with a neural representation  ... 
arXiv:1905.10679v4 fatcat:cl4wyoqtd5amdfcqlg5njzsd3i

Neural Ranking Models for Document Retrieval [article]

Mohamed Trabelsi, Zhiyu Chen, Brian D. Davison, Jeff Heflin
2021 arXiv   pre-print
A variety of deep learning models have been proposed, and each model presents a set of neural network components to extract features that are used for ranking.  ...  In our discussion of the literature, we analyze the promising neural components, and propose future research directions.  ...  Graph neural networks: A review of methods and applications. arXiv: 1812. 08434. Zhou, W., Li, H., & Tian, Q. (2017).  ... 
arXiv:2102.11903v1 fatcat:zc2otf456rc2hj6b6wpcaaslsa

Chinese Medical Question Answer Matching Based on Interactive Sentence Representation Learning [article]

Xiongtao Cui, Jungang Han
2020 arXiv   pre-print
inside the sentence and the semantic association between question and answer, and then combine with the multi-scale CNNs network or BiGRU network to take the advantage of different structure of neural  ...  To better adapt to Chinese medical question-answer matching and take the advantages of different neural network structures, we propose the Crossed BERT network to extract the deep semantic information  ...  MODELS The similarity between the sentence representations of questions and answers can measure their matching relationship, but it is only limited in the semantic information inside the sentence.  ... 
arXiv:2011.13573v1 fatcat:ywed62aegrfxnb5pq3ux2ymyiu

Content-based Representations of audio using Siamese neural networks [article]

Pranay Manocha, Rohan Badlani, Anurag Kumar, Ankit Shah, Benjamin Elizalde, Bhiksha Raj
2018 arXiv   pre-print
We propose a novel approach which encodes the audio into a vector representation using Siamese Neural Networks.  ...  Using simple similarity measures such as those based on simple euclidean distance and cosine similarity we show that these representations can be very effectively used for retrieving recordings similar  ...  We learn these semantic representations through Siamese Neural Networks. Fig 1 shows our proposed framework. A Siamese neural network actually consists of two twin networks.  ... 
arXiv:1710.10974v3 fatcat:uet2joclkfahzpdebzybavbx2m

Contrastive Similarity Matching for Supervised Learning [article]

Shanshan Qin, Nayantara Mudur, Cengiz Pehlevan
2020 arXiv   pre-print
We formulate this idea using a contrastive similarity matching objective function and derive from it deep neural networks with feedforward, lateral, and feedback connections, and neurons that exhibit biologically-plausible  ...  In both, representations of objects in the same category become progressively more similar, while objects belonging to different categories become less similar.  ...  We thank Dina Obeid and Blake Bordelon for helpful discussions.  ... 
arXiv:2002.10378v5 fatcat:vi7p4mejdjc23onelasln2zffm

Toward a Deep Neural Approach for Knowledge-Based IR [article]

Gia-Hung Nguyen, Lynda Tamine, Laure Soulier, Nathalie Bricon-Souf
2016 arXiv   pre-print
With this in mind, we argue that embedding KBs within deep neural architectures supporting documentquery matching would give rise to fine-grained latent representations of both words and their semantic  ...  In this paper, we review the main approaches of neural-based document ranking as well as those approaches for latent representation of entities and relations via KBs.  ...  More close to our work, we consider in this paper the specific task of text matching and the use of deep neural networks for document ranking.  ... 
arXiv:1606.07211v1 fatcat:jdypcyno3zcwphnoclk44dsfxi

Sparse Over-complete Patch Matching [article]

Akila Pemasiri and Kien Nguyen and Sridha Sridharan and Clinton Fookes
2018 arXiv   pre-print
State -of-the-art patch matching techniques take image patches as input to a convolutional neural network to extract the patch features and evaluate their similarity.  ...  We propose a new paradigm which encodes image patch details by encoding the patch and subsequently using this sparse representation as input to a neural network to compare the patches.  ...  In our approach we use coefficient resulted from over-complete sparse coding of patches ( Figure 1 , Figure 2 ) as the feature representation and to estimate the patch similarity we use a neural network  ... 
arXiv:1806.03556v2 fatcat:fztcsxckanakli74tp2w4gwda4

From Reality to Perception: Genre-Based Neural Image Style Transfer

Zhuoqi Ma, Nannan Wang, Xinbo Gao, Jie Li
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
We hope that this work provides new insight for including artists' perceptions into neural style transfer process, and helps people to understand the underlying characters of the artist or the genre.  ...  We collect a set of Van Gogh's paintings and cubist artworks, and their semantically corresponding real world photos.  ...  IAGR 20170103), in part by the Leading Talent of Technological Innovation of Ten-Thousands Talents Program under Grant CS31117200001.  ... 
doi:10.24963/ijcai.2018/485 dblp:conf/ijcai/MaWGL18 fatcat:iknfotjzqvgoleoslxltyn3u4q
« Previous Showing results 1 — 15 out of 204,314 results