A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Filters
Semi-automatic Technology Roadmapping Composing Method for Multiple Science, Technology, and Innovation Data Incorporation
[chapter]
2016
Innovation, Technology, and Knowledge Management
Preserved Fulltext
The understanding gained will assist in description of computer science macro-trends for R&D decision makers. ...
Since its first engagement with industry decades ago, Technology Roadmapping (TRM) is taking a more and more important role for technical intelligence in current R&D planning and innovation tracking. ...
We then apply Term Clumping steps for feature extraction, the process of which is given in Table 1 . in 2013 as a sample; also, we apply Pruning to DII first, and then, Fuzzy Matching. * #T = Number of ...
doi:10.1007/978-3-319-39056-7_12
fatcat:2zfhh4ju4fb65hchzgb6etuehu
Small Sample Learning in Big Data Era
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
As a promising area in artificial intelligence, a new learning paradigm, called Small Sample Learning (SSL), has been attracting prominent research attention in the recent years. ...
This category mainly focuses on learning with insufficient samples, and can also be called small data learning in some literatures. ...
( a )Figure 12 : a12 Siamese Networks for Chromosome Classification (b) Matching Network for Low Data Drug Discovery Examples of metric learning in SSL. ...
arXiv:1808.04572v3
fatcat:lqqzzrmgfnfb3izctvdzgopuny
Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-identification
2018
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Specifically, we introduce an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identitydiscriminative feature representation space transferrable ...
to any new (unseen) target domain for re-id tasks without the need for collecting new labelled training data from the target domain (i.e. unsupervised learning in the target domain). ...
Acknowledgements This work was partially supported by the China Scholarship Council, Vision Semantics Ltd, Royal Society Newton Advanced Fellowship Programme (NA150459), and In-novateUK Industrial Challenge ...
doi:10.1109/cvpr.2018.00242
dblp:conf/cvpr/WangZGL18
fatcat:6t46zlmqendb5i2fbuorvsrcw4
Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-Identification
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Specifically, we introduce an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identitydiscriminative feature representation space transferrable ...
to any new (unseen) target domain for re-id tasks without the need for collecting new labelled training data from the target domain (i.e. unsupervised learning in the target domain). ...
Acknowledgements This work was partially supported by the China Scholarship Council, Vision Semantics Ltd, Royal Society Newton Advanced Fellowship Programme (NA150459), and In-novateUK Industrial Challenge ...
arXiv:1803.09786v1
fatcat:zgdruvntp5f7feds5czgfvh7wa
Multi-View Transformer for 3D Visual Grounding
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose a Multi-View Transformer (MVT) for 3D visual grounding. ...
The multi-view space enables the network to learn a more robust multi-modal representation for 3D visual grounding and eliminates the dependence on specific views. ...
In the second stage, the task is modeled as a matching problem where 3D visual features are fused with the text features of the query language description to predict the matching scores. ...
arXiv:2204.02174v1
fatcat:2nmvi5o4gjex5dqi7b7fawtayu
Mining for Information Discovery on the Web: Overview and Illustrative Research
[chapter]
2004
Intelligent Technologies for Information Analysis
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
The Web has become a fertile ground for numerous research activities in mining. In this chapter we discuss research on finding targeted information on the Web. ...
We conclude by briefly discussing novel research opportunities in the area of mining for information discovery on the Web. ...
Acknowledgments: We thank Ying Lu and Yoonkyong Lee for obtaining the datasets and conducting the experiments in the PROM project, and for valuable comments on parts of this chapter. ...
doi:10.1007/978-3-662-07952-2_7
fatcat:kyp6jwkrl5dsvpl2f7l6b34j3a
A survey of joint intent detection and slot-filling models in natural language understanding
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We observe three milestones in this research so far: Intent detection to identify the speaker's intention, slot filling to label each word token in the speech/text, and finally, joint intent classification ...
Intent classification and slot filling are two critical tasks for natural language understanding. Traditionally the two tasks have been deemed to proceed independently. ...
SLU starts with automatic speech recognition (ASR), the task of taking the sound waves or images of expressed language, and transcribing to text. ...
arXiv:2101.08091v3
fatcat:ai6w2imilrfupf4m5fm2rjtzxi
Deep Learning for SAR Ship Detection: Past, Present and Future
2022
Remote Sensing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In the future part, we list the problem and direction of this field. We can find that, in the past five years, the AP50 has boosted from 78.8% in 2017 to 97.8 % in 2022 on SSDD. ...
After the revival of deep learning in computer vision in 2012, SAR ship detection comes into the deep learning era too. ...
[95] proposed a feature balance and matching network, which uses the Feature fusion. Chen et al. ...
doi:10.3390/rs14112712
fatcat:dbd6a4ugwjc65pook3wpcuj52a
Towards Robust Pattern Recognition: A Review
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
directions for robust pattern recognition. ...
The accuracies for many pattern recognition tasks have increased rapidly year by year, achieving or even outperforming human performance. ...
automatically learned and optimized fusion structure. ...
arXiv:2006.06976v1
fatcat:mn35i7bmhngl5hxr3vukdcmmde
A Review of Single-Source Deep Unsupervised Visual Domain Adaptation
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, we review the latest single-source deep unsupervised domain adaptation methods focused on visual tasks and discuss new perspectives for future research. ...
Large-scale labeled training datasets have enabled deep neural networks to excel across a wide range of benchmark vision tasks. ...
[69] recently proposed Transferrable Prototypical Networks, which perform domain alignment such that prototypes for each class in the source and target domains are close in the embedding space and the ...
arXiv:2009.00155v3
fatcat:yqkew4n4q5gtbjosozufw37ome
Language-Independent Tokenisation Rivals Language-Specific Tokenisation for Word Similarity Prediction
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Moreover, we find that smoothed inverse frequency (SIF) to be an accurate method to create word embeddings from subword embeddings for multilingual semantic similarity prediction tasks. ...
In this paper, we empirically compare the two approaches using semantic similarity measurement as an evaluation task across a diverse set of languages. ...
Given that LIT methods have been popularly used in NMT, where decoder vocabularies are typically less than 100K, it is encouraging to see that this benefit is transferrable to other NLP tasks such as semantic ...
arXiv:2002.11004v1
fatcat:2eplhu2a6zg23ejxyakak3o2de
A Hierarchical Network for Abstractive Meeting Summarization with Cross-Domain Pretraining
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Meanwhile, there are a handful of deep neural models for text summarization and dialogue systems. ...
In this paper, we propose a novel abstractive summary network that adapts to the meeting scenario. ...
Acknowledgement We thank William Hinthorn for proof-reading this paper. We thank the anonymous reviewers for their valuable comments. ...
arXiv:2004.02016v4
fatcat:yjbxv5h4lzapvaax2holwqp6ka
Book of Abstracts of the Digital Humanities in the Nordic Countries 5th conference. Riga, 20–23 October 2020
[article]
2020
Zenodo
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Book of Abstracts DHN, Rīga 2020 Book of Abstracts of the Digital Humanities in the Nordic Countries 5th conference. ...
Acknowledgements We thank the Fritz Thyssen Foundation for their funding for the research project Distant Viewing. ...
We thank the Staatsbibliothek Berlin for providing access to the Wegehaupt collection. ...
doi:10.5281/zenodo.4107117
fatcat:6ongky6p5rab7gvtawnjmp2ofm
Artificial Intelligence for the Metaverse: A Survey
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Finally, we conclude the key contribution of this survey and open some future research directions in AI for the metaverse. ...
We then convey a comprehensive investigation of AI-based methods concerning six technical aspects that have potentials for the metaverse: natural language processing, machine vision, blockchain, networking ...
Various CNNs and LSTM networks with simple structures and advanced-designed architectures. [31] Generate short text in image captioning and long text in virtual question answer. ...
arXiv:2202.10336v1
fatcat:35isd745dbaqfnpzthnmbaosue
A question-entailment approach to question answering
2019
BMC Bioinformatics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
First, we compare logistic regression and deep learning methods for RQE using different kinds of datasets including textual inference, question similarity, and entailment in both the open and clinical ...
Conclusions: The evaluation results support the relevance of question entailment for QA and highlight the effectiveness of combining IR and RQE for future QA efforts. ...
Shooshan (NLM/NIH) for her help with the judgment of the retrieved answers, and Ellen Voorhees (NIST) for her help with the TREC LiveQA evaluation. ...
doi:10.1186/s12859-019-3119-4
fatcat:ztrn5jaiwjdizmicuscn466ghq
« Previous
Showing results 1 — 15 out of 156 results