{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,8]],"date-time":"2026-02-08T07:53:45Z","timestamp":1770537225924,"version":"3.49.0"},"reference-count":30,"publisher":"China Science Publishing & Media Ltd.","issue":"3","license":[{"start":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T00:00:00Z","timestamp":1645660800000},"content-version":"vor","delay-in-days":54,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2022,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Few-shot learning has been proposed and rapidly emerging as a viable means for completing various tasks. Many few-shot models have been widely used for relation learning tasks. However, each of these models has a shortage of capturing a certain aspect of semantic features, for example, CNN on long-range dependencies part, Transformer on local features. It is difficult for a single model to adapt to various relation learning, which results in a high variance problem. Ensemble strategy could be competitive in improving the accuracy of few-shot relation extraction and mitigating high variance risks. This paper explores an ensemble approach to reduce the variance and introduces fine-tuning and feature attention strategies to calibrate relation-level features. Results on several few-shot relation learning tasks show that our model significantly outperforms the previous state-of-the-art models.<\/jats:p>","DOI":"10.1162\/dint_a_00144","type":"journal-article","created":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T16:29:21Z","timestamp":1645720161000},"page":"529-551","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":15,"title":["Ensemble Making Few-Shot Learning Stronger"],"prefix":"10.3724","volume":"4","author":[{"given":"Qiang","family":"Lin","sequence":"first","affiliation":[{"name":"Computer School, University of South China 42,1001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yongbin","family":"Liu","sequence":"additional","affiliation":[{"name":"Computer School, University of South China 42,1001, China"},{"name":"Hunan provincial base for scientific and technological innovation cooperation, Hunan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wen","family":"Wen","sequence":"additional","affiliation":[{"name":"Computer School, University of South China 42,1001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhihua","family":"Tao","sequence":"additional","affiliation":[{"name":"Computer School, University of South China 42,1001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chunping","family":"Ouyang","sequence":"additional","affiliation":[{"name":"Computer School, University of South China 42,1001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yaping","family":"Wan","sequence":"additional","affiliation":[{"name":"Computer School, University of South China 42,1001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"2026","published-online":{"date-parts":[[2022,7,1]]},"reference":[{"key":"2022081019142947100_ref1","article-title":"Siamese neural networks for one-shot image\n                        recognition","volume":"2","author":"Koch","year":"2015","journal-title":"In: ICML Deep Learning\n                        Workshop,"},{"key":"2022081019142947100_ref2","article-title":"Matching networks for one shot learning","volume":"29","author":"Vinyals","year":"2016","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2022081019142947100_ref3","first-page":"1199","volume-title":"Learning to compare: Relation network for few-shot\n                    learning","author":"Sung","year":"2018"},{"key":"2022081019142947100_ref4","article-title":"Prototypical networks for few-shot\n                        learning[J]","volume":"30","author":"Snell","year":"2017","journal-title":"Advances in Neural Information\n                        Processing Systems"},{"key":"2022081019142947100_ref5","volume-title":"A baseline for few-shot image classification","author":"Dhillon","year":"2019"},{"key":"2022081019142947100_ref6","first-page":"3723","volume-title":"Diversity with cooperation: Ensemble methods for few-shot\n                        classification","author":"Dvornik","year":"2019"},{"issue":"4","key":"2022081019142947100_ref7","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1162\/neco.1989.1.4.541","article-title":"Backpropagation applied to handwritten zip code\n                        recognition","volume":"1","author":"Le Cun","year":"1989","journal-title":"Neural Computation"},{"key":"2022081019142947100_ref8","first-page":"1","volume-title":"Going deeper with convolutions","author":"Szegedy","year":"2015"},{"key":"2022081019142947100_ref9","volume-title":"Learning phrase representations using RNN encoderdecoder for\n                        statistical machine translation","author":"Cho","year":"2014"},{"key":"2022081019142947100_ref10","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Advances\n                        in Neural Information Processing Systems"},{"key":"2022081019142947100_ref11","volume-title":"A new meta-baseline for few-shot learning","author":"Chen","year":"2020"},{"key":"2022081019142947100_ref12","volume-title":"A closer look at few-shot classification","author":"Chen","year":"2019"},{"key":"2022081019142947100_ref13","volume-title":"Fewrel: A large-scale supervised few-shot relation classification\n                        dataset with state-of-the-art evaluation","author":"Han","year":"2018"},{"key":"2022081019142947100_ref14","volume-title":"FewRel 2.0: Towards more challenging few-shot relation\n                        classification","author":"Gao","year":"2019"},{"key":"2022081019142947100_ref15","first-page":"2554","volume-title":"Meta\n                        networks","author":"Munkhdalai","year":"2020"},{"key":"2022081019142947100_ref16","volume-title":"Optimization as a model for few-shot learning","author":"Ravi","year":"2016"},{"key":"2022081019142947100_ref17","volume-title":"Meta-learning: A survey","author":"Vanschoren","year":"2018"},{"key":"2022081019142947100_ref18","first-page":"12365","volume-title":"Meta-learning of neural architectures for few-shot\n                        learning","author":"Elsken","year":"2020"},{"key":"2022081019142947100_ref19","first-page":"1126","volume-title":"Model-agnostic meta-learning for fast adaptation of deep\n                        networks","author":"Finn","year":"2017"},{"key":"2022081019142947100_ref20","article-title":"Bayesian model-agnostic meta-learning","volume":"31","author":"Yoon","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2022081019142947100_ref21","first-page":"7867","volume-title":"Few-shot relation extraction via bayesian meta-learning on relation\n                        graphs","author":"Qu","year":"2020"},{"key":"2022081019142947100_ref22","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/P19-1277","volume-title":"Multi-level matching and aggregation network for few-shot relation\n                        classification","author":"Ye","year":"2019"},{"issue":"01","key":"2022081019142947100_ref23","first-page":"6407","article-title":"Hybrid attention-based prototypical networks for noisy\n                        few-shot relation classification","volume":"33","author":"Gao","year":"2019","journal-title":"In: Proceedings of\n                        the AAAI Conference on Artificial Intelligence,"},{"key":"2022081019142947100_ref24","volume-title":"Bert: Pre-training of deep bidirectional transformers for language\n                        understanding","author":"Devlin","year":"2018"},{"key":"2022081019142947100_ref25","volume-title":"Improving language understanding by generative pretraining","author":"Radford","year":"2018"},{"key":"2022081019142947100_ref26","doi-asserted-by":"crossref","first-page":"114135","DOI":"10.1016\/j.psychres.2021.114135","article-title":"Detecting formal thought disorder by deep contextualized word\n                        representations","volume":"304","author":"Sarzynska-Wawer","year":"2021","journal-title":"Psychiatry Research"},{"key":"2022081019142947100_ref27","doi-asserted-by":"crossref","DOI":"10.24963\/ijcai.2018\/630","volume-title":"Ensemble neural relation extraction with adaptive boosting","author":"Yang","year":"2018"},{"key":"2022081019142947100_ref28","volume-title":"FewRel 2.0: Towards more challenging few-shot relation\n                        classification","author":"Gao","year":"2019"},{"key":"2022081019142947100_ref29","volume-title":"Few-shot learning with graph neural networks","author":"Garcia","year":"2017"},{"key":"2022081019142947100_ref30","volume-title":"A simple neural attentive meta-learner[J]","author":"Mishra","year":"2017"}],"container-title":["Data Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/dint\/article-pdf\/4\/3\/529\/2038454\/dint_a_00144.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/dint\/article-pdf\/4\/3\/529\/2038454\/dint_a_00144.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,14]],"date-time":"2025-03-14T07:44:02Z","timestamp":1741938242000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.sciengine.com\/doi\/10.1162\/dint_a_00144"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022]]},"references-count":30,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2022,7,1]]}},"URL":"https:\/\/doi.org\/10.1162\/dint_a_00144","relation":{},"ISSN":["2641-435X"],"issn-type":[{"value":"2641-435X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022]]},"published":{"date-parts":[[2022]]}}}