Filters








344,599 Hits in 4.9 sec

Multi-modal dialog for browsing large visual catalogs using exploration-exploitation paradigm in a joint embedding space [article]

Indrani Bhattacharya, Arkabandhu Chowdhury, Vikas Raykar
2019 arXiv   pre-print
Our system remembers the context of the dialog and uses an exploration-exploitation paradigm to assist in visual browsing.  ...  We formulate our problem of "showing k best images to a user" based on the dialog context so far, as sampling from a Gaussian Mixture Model in a high dimensional joint multi-modal embedding space, that  ...  However, this architecture cannot learn disentangled representations of the different attributes associated with a product.  ... 
arXiv:1901.09854v2 fatcat:mccrz4unfjhqbex75iq6eisody

ANRL: Attributed Network Representation Learning via Deep Neural Networks

Zhen Zhang, Hongxia Yang, Jiajun Bu, Sheng Zhou, Pinggang Yu, Jianwei Zhang, Martin Ester, Can Wang
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
Network representation learning (RL) aims to transform the nodes in a network into low-dimensional vector spaces while preserving the inherent properties of the network.  ...  To capture the network structure, attribute-aware skip-gram model is designed based on the attribute encoder to formulate the correlations between each node and its direct or indirect neighbors.  ...  nodes with similar context should be similar in latent semantic space.  ... 
doi:10.24963/ijcai.2018/438 dblp:conf/ijcai/ZhangYBZYZE018 fatcat:7vqslkhb6jglhac6w7qmkcsykm

Understanding Art through Multi-Modal Retrieval in Paintings [article]

Noa Garcia, Benjamin Renoust, Yuta Nakashima
2019 arXiv   pre-print
representations in artistic images.  ...  We introduce the use of multi-modal techniques in the field of automatic art analysis by 1) collecting a multi-modal dataset with fine-art paintings and comments, and 2) exploring robust visual and textual  ...  Conclusion We addressed art understanding by introducing a new dataset of paintings with associated comments and explor- ing multi-modal representations in art.  ... 
arXiv:1904.10615v1 fatcat:7szy5ph7tbb43nxby7ocizgtwe

Interactions in Visualization

Maria Lujan Ganuza
2018 Journal of Computer Science and Technology  
The main contributions of this thesis are: • A specification of a classification for data, attributes and data sets in the visualization context.  ...  In this context, it is necessary to define a representation for the data sets involved in the process.  ...  In Spinel Explorer, the designed interactions greatly improve the process of classifying mineral samples.  ... 
doi:10.24215/16666038.18.e20 fatcat:hnbrj5uw4na5lexxuyc3pnocee

Exemplars-guided Empathetic Response Generation Controlled by the Elements of Human Communication [article]

Navonil Majumder, Deepanway Ghosal, Devamanyu Hazarika, Alexander Gelbukh, Rada Mihalcea, Soujanya Poria
2021 arXiv   pre-print
The majority of existing methods for empathetic response generation rely on the emotion of the context to generate empathetic responses.  ...  However, empathy is much more than generating responses with an appropriate emotion.  ...  The EmpathyMentalHealth contains (context, response) pairs with annotated emotional presence, interpretation, exploration labels.  ... 
arXiv:2106.11791v3 fatcat:ob4n5ztu7ndavm6tizpz7dpowm

ContextNet: representation and exploration for painting classification and retrieval in context

Noa Garcia, Benjamin Renoust, Yuta Nakashima
2019 International Journal of Multimedia Information Retrieval  
In this way, we are able to (1) capture information about the content and the style with the visual representations and (2) encode relationships between different artistic attributes with the ContextNet  ...  As context can be obtained from multiple sources, we explore two modalities of ContextNets: one based on multitask learning and another one based on knowledge graphs.  ...  Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long  ... 
doi:10.1007/s13735-019-00189-4 fatcat:hek35v7hlrevvp354advr76rf4

An Adversarial Learning Framework For A Persona-Based Multi-Turn Dialogue Model [article]

Oluwatobi Olabiyi, Anish Khazane, Alan Salimov, Erik T. Mueller
2019 arXiv   pre-print
We also explore the trade-offs from using either variant of phredGAN on datasets with many but weak attribute modalities (such as with Big Bang Theory and Friends) and ones with few but strong attribute  ...  We also explore two approaches to accomplish the conditional discriminator: (1) phredGAN_a, a system that passes the attribute representation as an additional input into a traditional adversarial discriminator  ...  Encoder: The context RNN, cRN N takes the source attribute c i as an additional input by concatenating its representation with the output of eRN N as in Figure 1 .  ... 
arXiv:1905.01992v2 fatcat:zhbfeptbybdshcbewp7vm57eqy

Weakly Supervised Learning by a Confusion Matrix of Contexts [chapter]

William Wu
2019 Lecture Notes in Computer Science  
reflected the Anna Karenina principle well -"Happy families are all alike; every unhappy family is unhappy in its own way", an encouraging sign to further explore contexts associated with harmonizing  ...  When the negative and positive labeled samples and misclassification errors are compared to "happy families" and "unhappy families", the contexts constructed by this model in the classification experiments  ...  While systematic sampling can be considered as a kind of stratification with equal sampling fractions, its value growth representation can be over-optimistic and will need improvement.  ... 
doi:10.1007/978-3-030-26142-9_6 fatcat:27tyndeqhjbcrksv2bfnfeybte

Attention-Guide Walk Model in Heterogeneous Information Network for Multi-Style Recommendation Explanation

Xin Wang, Ying Wang, Yunzhi Ling
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Inspired by the attention mechanism, we determine the important contexts for recommendation explanation and learn joint representation of multi-style user-item interactions for enhancing recommendation  ...  To address these issues, we propose a framework (MSRE) of generating the multi-style recommendation explanation with the attention-guide walk model on affiliation relations and interaction relations in  ...  In addition, joint representation of context information with the help of attention mechanism can greatly improve the performance of recommendation. Case Study.  ... 
doi:10.1609/aaai.v34i04.6095 fatcat:mb636x2jbbhgxhxjyzxihj53w4

Context Interchange Mediation for Semantic Interoperability and Dynamic Integration of Autonomous Information Sources in the Fixed Income Securities Industry

Allen Moulton, Stuart E. Madnick, Michael Siegel
2003 Social Science Research Network  
We examine semantic interoperability problems in the fixed income securities industry and propose a knowledge representation architecture for context interchange mediation to support dynamic integration  ...  includes: 1) data models for each source and receiver, 2) subject ontologies, containing abstract subject matter conceptualizations that would be known to experienced practitioners in the industry, and 3) context  ...  Context model fragment for R specify the data representation used in the context. Sources A and B would be modeled similarly. Source C would be modeled with a inmSecurity object.  ... 
doi:10.2139/ssrn.376863 fatcat:r3q4bi7bszci3lfx2hosul7sei

Knowledge Representation Architecture for Context Interchange Mediation: Fixed Income Securities Investment Examples

Allen Moulton, Stuart E. Madnick, Michael Siegel
2001 Social Science Research Network  
We examine a knowledge representation architecture to support context interchange mediation.  ...  data models for each source and receiver, 2) subject domain ontologies, containing abstract subject matter conceptualizations that would be known to experienced practitioners in the industry, and 3) context  ...  02-11-1992 issue date, MM-DD-YYYY firstCoup 08-01-1992 attribute sample data semantic notes 10yr 5.091 yield on current 10 year T-note attribute sample data semantic notes price 117.875  ... 
doi:10.2139/ssrn.289996 fatcat:jp6z6xvjpnd5zlbn5kfc45vuxi

DisCont: Self-Supervised Visual Attribute Disentanglement using Context Vectors [article]

Sarthak Bhagat, Vishaal Udandarao, Shagun Uppal
2020 arXiv   pre-print
Disentangling the underlying feature attributes within an image with no prior supervision is a challenging task.  ...  Models that can disentangle attributes well provide greater interpretability and control.  ...  We construct context vectors by aggregating all the i th feature attributes locally within the sampled minibatch.  ... 
arXiv:2006.05895v2 fatcat:bqml26d7ozb6dpi64zm5mqjqtm

Learning Geo-Temporal Image Features [article]

Menghua Zhai and Tawfiq Salem and Connor Greenwell and Scott Workman and Robert Pless and Nathan Jacobs
2019 arXiv   pre-print
In this section, we explore the ability of this representation for directly estimating transient attributes.  ...  correlations with transient image attributes.  ... 
arXiv:1909.07499v1 fatcat:mfhaprtepnb7npd3jmliurwuwi

Exploring synonyms as context in zero-shot action recognition

Ioannis Alexiou, Tao Xiang, Shaogang Gong
2016 2016 IEEE International Conference on Image Processing (ICIP)  
In this work we propose to explore a broader semantic contextual information in the text domain to enrich the word vector representation of action classes.  ...  Moreover, it also outperforms attribute embedding ZSL with human annotation.  ...  Here we wish to explore a broader semantic context of the target action class name.  ... 
doi:10.1109/icip.2016.7533149 dblp:conf/icip/AlexiouXG16 fatcat:xt3ipbrlmjbubchaoawxthjmy4

CA3Net: Contextual-Attentional Attribute-Appearance Network for Person Re-Identification [article]

Jiawei Liu and Zheng-Jun Zha and Hongtao Xie and Zhiwei Xiong and Yongdong Zhang
2018 arXiv   pre-print
Specifically, an attribute network within CA3Net is designed with an Attention-LSTM module.  ...  It concentrates the network on latent image regions related to each attribute as well as exploits the semantic context among attributes by a LSTM module.  ...  The proposed CA 3 Net jointly learns semantic attribute and visual appearance representation of pedestrians with simultaneous exploration of the semantic context among attributes, visual attention on attributes  ... 
arXiv:1811.07544v1 fatcat:qsgoqomymjbuzmqe3wk3nljarm
« Previous Showing results 1 — 15 out of 344,599 results