File Download
Supplementary
-
Citations:
- Appears in Collections:
Conference Paper: Large margin meta-learning for few-shot classification
Title | Large margin meta-learning for few-shot classification |
---|---|
Authors | |
Issue Date | 2018 |
Publisher | Neural Information Processing Systems Foundation. |
Citation | The 2nd Workshop on Meta-Learning at the 32nd International Conference on Neural Information Processing Systems (NeurIPS) 2018, Montreal, Canada, 3-8 December 2018 How to Cite? |
Abstract | This paper proposes a large margin principle to overcome the problem of data scarcity in training meta-learning models for few-shot classification. To realize it, we develop an efficient framework that can be easily incorporated in existing metric-based meta-learning models. We demonstrate that the large margin principle can improve the generalization capacity of state-of-the-art meta-learning methods and lead to consistent and considerable improvements on few-shot classification. |
Description | Spotlights 1 |
Persistent Identifier | http://hdl.handle.net/10722/278330 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Wang, Y | - |
dc.contributor.author | Wu, XM | - |
dc.contributor.author | Li, Q | - |
dc.contributor.author | Gu, J | - |
dc.contributor.author | Xiang, W | - |
dc.contributor.author | Zhang, L | - |
dc.contributor.author | Li, VOK | - |
dc.date.accessioned | 2019-10-04T08:11:55Z | - |
dc.date.available | 2019-10-04T08:11:55Z | - |
dc.date.issued | 2018 | - |
dc.identifier.citation | The 2nd Workshop on Meta-Learning at the 32nd International Conference on Neural Information Processing Systems (NeurIPS) 2018, Montreal, Canada, 3-8 December 2018 | - |
dc.identifier.uri | http://hdl.handle.net/10722/278330 | - |
dc.description | Spotlights 1 | - |
dc.description.abstract | This paper proposes a large margin principle to overcome the problem of data scarcity in training meta-learning models for few-shot classification. To realize it, we develop an efficient framework that can be easily incorporated in existing metric-based meta-learning models. We demonstrate that the large margin principle can improve the generalization capacity of state-of-the-art meta-learning methods and lead to consistent and considerable improvements on few-shot classification. | - |
dc.language | eng | - |
dc.publisher | Neural Information Processing Systems Foundation. | - |
dc.relation.ispartof | 32nd International Conference on Neural Information Processing Systems (NeurIPS 2018) - Workshop on Meta-Learning (MetaLearn 2018) | - |
dc.relation.ispartof | Workshop on Meta-Learning (MetaLearn 2018) @ NeurIPS 2018 | - |
dc.title | Large margin meta-learning for few-shot classification | - |
dc.type | Conference_Paper | - |
dc.identifier.email | Li, VOK: vli@eee.hku.hk | - |
dc.identifier.authority | Li, VOK=rp00150 | - |
dc.description.nature | published_or_final_version | - |
dc.identifier.hkuros | 306532 | - |
dc.publisher.place | Canada | - |