Filters








14,263 Hits in 4.1 sec

Meta-Transfer Learning for Few-Shot Learning [article]

Qianru Sun, Yaoyao Liu, Tat-Seng Chua, Bernt Schiele
<span title="2019-04-09">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper we propose a novel few-shot learning method called meta-transfer learning (MTL) which learns to adapt a deep NN for few shot learning tasks.  ...  Meta-learning has been proposed as a framework to address the challenging few-shot learning setting.  ...  Few-shot learning tasks have been defined for this purpose. The aim is to learn new concepts from few labeled examples, e.g. 1-shot learning [25] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1812.02391v3">arXiv:1812.02391v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/mifgrgaqbramfmk3xshkqqlkla">fatcat:mifgrgaqbramfmk3xshkqqlkla</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825200336/https://arxiv.org/pdf/1812.02391v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/0a/3a/0a3a003457f5d7758a42a0e4b7278b39a86ed0bd.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1812.02391v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Meta-Transfer Learning for Few-Shot Learning

Qianru Sun, Yaoyao Liu, Tat-Seng Chua, Bernt Schiele
<span title="">2019</span> <i title="IEEE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ilwxppn4d5hizekyd3ndvy2mii" style="color: black;">2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)</a> </i> &nbsp;
In this paper we propose a novel few-shot learning method called meta-transfer learning (MTL) which learns to adapt a deep NN for few shot learning tasks.  ...  Meta-learning has been proposed as a framework to address the challenging few-shot learning setting.  ...  Few-shot learning tasks have been defined for this purpose.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/cvpr.2019.00049">doi:10.1109/cvpr.2019.00049</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/cvpr/SunLCS19.html">dblp:conf/cvpr/SunLCS19</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/d27j662prfglnoarnnz3c5ziy4">fatcat:d27j662prfglnoarnnz3c5ziy4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200320090940/https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=5450&amp;context=sis_research" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/0d/09/0d09fffe4ca9dbde56d3af4650d4bae910425bac.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/cvpr.2019.00049"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels [article]

Massimiliano Patacchiola, Jack Turner, Elliot J. Crowley, Michael O'Boyle, Amos Storkey
<span title="2020-10-13">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We empirically demonstrate that DKT outperforms several state-of-the-art algorithms in few-shot classification, and is the state of the art for cross-domain adaptation and regression.  ...  Following the recognition that meta-learning is implementing learning in a multi-level model, we present a Bayesian treatment for the meta-learning inner loop through the use of deep kernels.  ...  Meta-learning (Bengio et al., 1992; Schmidhuber, 1992; Hospedales et al., 2020) methods have become very popular for few-shot learning tasks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.05199v4">arXiv:1910.05199v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/otdhlal4qzfzxhbnzdwnhpg7ma">fatcat:otdhlal4qzfzxhbnzdwnhpg7ma</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201015124707/https://arxiv.org/pdf/1910.05199v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/2c/59/2c59e5d5abe312b1ae6a69bbc4a719d8e565779c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.05199v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

SB-MTL: Score-based Meta Transfer-Learning for Cross-Domain Few-Shot Learning [article]

John Cai, Bill Cai, Sheng Mei Shen
<span title="2020-12-03">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
in Cross-Domain Few-Shot Learning (CD-FSL).  ...  Our method, called Score-based Meta Transfer-Learning (SB-MTL), combines transfer-learning and meta-learning by using a MAML-optimized feature encoder and a score-based Graph Neural Network.  ...  Conclusion In this paper, we have proposed Score-Based Meta Transfer-Learning to address the Cross-Domain Few-Shot Learning problem.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.01784v1">arXiv:2012.01784v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/njk5flv7jndwdmouibf2di4x7i">fatcat:njk5flv7jndwdmouibf2di4x7i</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201206044703/https://arxiv.org/pdf/2012.01784v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c4/57/c4575608bcb1aa459baf5a1e19817aa37385b60c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.01784v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Meta-Voice: Fast few-shot style transfer for expressive voice cloning using meta learning [article]

Songxiang Liu, Dan Su, Dong Yu
<span title="2021-11-14">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we approach to the hard fast few-shot style transfer for voice cloning task using meta learning.  ...  We investigate the model-agnostic meta-learning (MAML) algorithm and meta-transfer a pre-trained multi-speaker and multi-prosody base TTS model to be highly sensitive for adaptation with few samples.  ...  The task of fast few-shot style transfer is very challenging in the sense that the learning algorithm needs to deal with not only a few-shot voice cloning problem (i.e., cloning a new voice using few samples  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.07218v1">arXiv:2111.07218v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ibb34g7huzbeppe4vkhgha6hx4">fatcat:ibb34g7huzbeppe4vkhgha6hx4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211130214629/https://arxiv.org/pdf/2111.07218v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/10/7a/107af906c158c021d9fca24bd135fcc551c0e662.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.07218v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

A Meta-Learning Approach for Custom Model Training [article]

Amir Erfan Eshratifar, Mohammad Saeed Abrishami, David Eigen, Massoud Pedram
<span title="2019-02-08">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In few-class, few-shot target task settings (i.e. when there are only a few classes and training examples available in the target task), meta-learning approaches that optimize for future task learning  ...  Transfer-learning and meta-learning are two effective methods to apply knowledge learned from large data sources to new tasks.  ...  Note that for many-classes (35-ways) tasks, the transfer learning baseline outperforms previous meta-learning algorithms, while in few-classes problems, the result is reversed: meta-learning beats transfer  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1809.08346v2">arXiv:1809.08346v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/57jjm3rrhvd5dle5mvsdigkjti">fatcat:57jjm3rrhvd5dle5mvsdigkjti</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825215829/https://arxiv.org/pdf/1809.08346v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/b8/f7/b8f738abd59912db7c71cb5ba93b47283345b0fc.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1809.08346v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

A Meta-Learning Approach for Custom Model Training

Amir Erfan Eshratifar, Mohammad Saeed Abrishami, David Eigen, Massoud Pedram
<span title="2019-07-17">2019</span> <i title="Association for the Advancement of Artificial Intelligence (AAAI)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/wtjcymhabjantmdtuptkk62mlq" style="color: black;">PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE</a> </i> &nbsp;
In few-class, few-shot target task settings (i.e. when there are only a few classes and training examples available in the target task), meta-learning approaches that optimize for future task learning  ...  Transfer-learning and meta-learning are two effective methods to apply knowledge learned from large data sources to new tasks.  ...  Note that for many-classes (35-ways) tasks, the transfer learning baseline outperforms previous meta-learning algorithms, while in few-classes problems, the result is reversed: meta-learning beats transfer  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1609/aaai.v33i01.33019937">doi:10.1609/aaai.v33i01.33019937</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hfk7jevqpbfpjoim6zvij3uauy">fatcat:hfk7jevqpbfpjoim6zvij3uauy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200305174257/https://aaai.org/ojs/index.php/AAAI/article/download/5105/4978" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/2d/60/2d605b3a648f873ef1e8961e30c0e85641b6b193.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1609/aaai.v33i01.33019937"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning [article]

Zhiqiang Shen and Zechun Liu and Jie Qin and Marios Savvides and Kwang-Ting Cheng
<span title="2021-02-08">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
to novel data, i.e. learning to transfer in few-shot scenario.) or meta-learning.  ...  The goal of few-shot learning is to learn a classifier that can recognize unseen classes from limited support data with labels.  ...  Here we also review the searching based methods for few-shot learning in this section. Meta-based few-shot learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2102.03983v1">arXiv:2102.03983v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/u37noahnpze4jm6ba4pcqihdyu">fatcat:u37noahnpze4jm6ba4pcqihdyu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210213202911/https://arxiv.org/pdf/2102.03983v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/7b/b3/7bb3a96078c854f03d6e9c2032ad6efe6873d974.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2102.03983v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Few-Shot Rolling Bearing Fault Diagnosis with Metric-Based Meta Learning

Sihan Wang, Dazhi Wang, Deshan Kong, Jiaxing Wang, Wenhui Li, Shuai Zhou
<span title="2020-11-11">2020</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/taedaf6aozg7vitz5dpgkojane" style="color: black;">Sensors</a> </i> &nbsp;
Therefore, in this paper, we propose a metric-based meta-learning method named Reinforce Relation Network (RRN) for diagnosing bearing faults with few-shot samples.  ...  The comparison studies with other methods demonstrate the superiority of the proposed method in the few-shot scenario.  ...  • A metric-based, few-shot, meta-learning framework is designed for bearing fault diagnosis, which is more suitable for a few-shot transfer scenario from the experimental situation to the actual working  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/s20226437">doi:10.3390/s20226437</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/33187173">pmid:33187173</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ilow3gnaavd6jhpkre3nyyxpui">fatcat:ilow3gnaavd6jhpkre3nyyxpui</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201113012152/https://res.mdpi.com/d_attachment/sensors/sensors-20-06437/article_deploy/sensors-20-06437.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/73/e8/73e81bf73c0eb1236c872ef2d377a2b345708eec.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/s20226437"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> mdpi.com </button> </a>

Meta-Transfer Learning through Hard Tasks [article]

Qianru Sun, Yaoyao Liu, Zhaozheng Chen, Tat-Seng Chua, Bernt Schiele
<span title="2019-10-07">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we propose a novel approach called meta-transfer learning (MTL) which learns to transfer the weights of a deep NN for few-shot learning tasks.  ...  Meta-learning has been proposed as a framework to address the challenging few-shot learning setting.  ...  [46] presented a meta-learning approach with convex base-learners for few-shot tasks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.03648v1">arXiv:1910.03648v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/l2z7dowb5bclzgr2a3ofk3z2za">fatcat:l2z7dowb5bclzgr2a3ofk3z2za</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200830052600/https://arxiv.org/pdf/1910.03648v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/0b/ec/0beceb58bf35073bf4abe1e0d9116c8525530179.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.03648v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

A Broader Study of Cross-Domain Few-Shot Learning [article]

Yunhui Guo, Noel C. Codella, Leonid Karlinsky, James V. Codella, John R. Smith, Kate Saenko, Tajana Rosing, Rogerio Feris
<span title="2020-07-17">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Extensive experiments on the proposed benchmark are performed to evaluate state-of-art meta-learning approaches, transfer learning approaches, and newer methods for cross-domain few-shot learning.  ...  Recent progress on few-shot learning largely relies on annotated data for meta-learning: base classes sampled from the same domain as the novel classes.  ...  Another line of work for few-shot learning uses a broader variety of classifiers for transfer learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1912.07200v3">arXiv:1912.07200v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nn5ls5ww3belvkc46jnhjwamyu">fatcat:nn5ls5ww3belvkc46jnhjwamyu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200806164951/https://arxiv.org/pdf/1912.07200v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1912.07200v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Meta Relational Learning for Few-Shot Link Prediction in Knowledge Graphs [article]

Mingyang Chen, Wen Zhang, Wei Zhang, Qiang Chen, Huajun Chen
<span title="2019-09-04">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We solve few-shot link prediction by focusing on transferring relation-specific meta information to make model learn the most important knowledge and learn faster, corresponding to relation meta and gradient  ...  In this work, we propose a Meta Relational Learning (MetaR) framework to do the common but challenging few-shot link prediction in KGs, namely predicting new triples about a relation by only observing  ...  Acknowledgments We want to express gratitude to the anonymous reviewers for their hard work and kind comments, which will further improve our work in the future.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.01515v1">arXiv:1909.01515v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/xvxmqx6bcbdu7l4qxwf5fzo3di">fatcat:xvxmqx6bcbdu7l4qxwf5fzo3di</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200824140741/https://arxiv.org/pdf/1909.01515v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/90/14/9014c4e7bec425a117b514e41d9cdf9dec8bd6c1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.01515v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Meta-Baseline: Exploring Simple Meta-Learning for Few-Shot Learning [article]

Yinbo Chen, Zhuang Liu, Huijuan Xu, Trevor Darrell, Xiaolong Wang
<span title="2021-08-19">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Meta-learning has been the most common framework for few-shot learning in recent years.  ...  The edge between these two lines of works has yet been underexplored, and the effectiveness of meta-learning in few-shot learning remains unclear.  ...  We thank Hang Gao for the helpful discussions.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.04390v4">arXiv:2003.04390v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/gq3baa5ugvgzhj235pvkbjkd6y">fatcat:gq3baa5ugvgzhj235pvkbjkd6y</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210824235725/https://arxiv.org/pdf/2003.04390v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/35/17/35172f0513d6a9fb9b1a40b7ad466da96a112d68.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.04390v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Meta Relational Learning for Few-Shot Link Prediction in Knowledge Graphs

Mingyang Chen, Wen Zhang, Wei Zhang, Qiang Chen, Huajun Chen
<span title="">2019</span> <i title="Association for Computational Linguistics"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/u3ideoxy4fghvbsstiknuweth4" style="color: black;">Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)</a> </i> &nbsp;
We solve few-shot link prediction by focusing on transferring relation-specific meta information to make model learn the most important knowledge and learn faster, corresponding to relation meta and gradient  ...  In this work, we propose a Meta Relational Learning (MetaR) framework to do the common but challenging few-shot link prediction in KGs, namely predicting new triples about a relation by only observing  ...  Acknowledgments We want to express gratitude to the anonymous reviewers for their hard work and kind comments, which will further improve our work in the future.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/d19-1431">doi:10.18653/v1/d19-1431</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/emnlp/ChenZZCC19.html">dblp:conf/emnlp/ChenZZCC19</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ac6ozrrovzft7grojqwza4dzwe">fatcat:ac6ozrrovzft7grojqwza4dzwe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200909060326/https://www.aclweb.org/anthology/D19-1431.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/4e/40/4e408b02fe3683d06ddc11c545c726961af83d58.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/d19-1431"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Meta-free few-shot learning via representation learning with weight averaging [article]

Kuilin Chen, Chi-Guhn Lee
<span title="2022-04-30">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recent studies on few-shot classification using transfer learning pose challenges to the effectiveness and efficiency of episodic meta-learning algorithms.  ...  To tackle the aforementioned issues, we propose a new transfer learning method to obtain accurate and reliable models for few-shot regression and classification.  ...  Therefore, the resulting method, called Meta-Free Representation Learning (MFRL), overcomes the aforementioned limitations in existing transfer learning methods for few-shot learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2204.12466v2">arXiv:2204.12466v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wspokfsrrjgpbewdgn5ha4offm">fatcat:wspokfsrrjgpbewdgn5ha4offm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220506182220/https://arxiv.org/pdf/2204.12466v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/db/0a/db0aefaed7f4e7d5826b885275d931638b1bb6eb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2204.12466v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 14,263 results