Unsupervised Feature Learning with Emergent Data-Driven Prototypicality

人工智能 模式识别(心理学) 双曲空间 数学 欧几里德距离 无监督学习 特征向量 特征(语言学) 一般化 欧几里得空间 计算机科学 几何学 组合数学 语言学 数学分析 哲学
作者
Yunhui Guo,Youren Zhang,Yubei Chen,Stella X. Yu
出处
期刊:Cornell University - arXiv
标识
DOI:10.48550/arxiv.2307.01421
摘要

Given an image set without any labels, our goal is to train a model that maps each image to a point in a feature space such that, not only proximity indicates visual similarity, but where it is located directly encodes how prototypical the image is according to the dataset. Our key insight is to perform unsupervised feature learning in hyperbolic instead of Euclidean space, where the distance between points still reflect image similarity, and yet we gain additional capacity for representing prototypicality with the location of the point: The closer it is to the origin, the more prototypical it is. The latter property is simply emergent from optimizing the usual metric learning objective: The image similar to many training instances is best placed at the center of corresponding points in Euclidean space, but closer to the origin in hyperbolic space. We propose an unsupervised feature learning algorithm in Hyperbolic space with sphere pACKing. HACK first generates uniformly packed particles in the Poincar\'e ball of hyperbolic space and then assigns each image uniquely to each particle. Images after congealing are regarded more typical of the dataset it belongs to. With our feature mapper simply trained to spread out training instances in hyperbolic space, we observe that images move closer to the origin with congealing, validating our idea of unsupervised prototypicality discovery. We demonstrate that our data-driven prototypicality provides an easy and superior unsupervised instance selection to reduce sample complexity, increase model generalization with atypical instances and robustness with typical ones.

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
1秒前
今后应助在下小绿采纳,获得10
1秒前
1秒前
电脑桌发布了新的文献求助10
1秒前
余空完成签到 ,获得积分10
2秒前
2秒前
DUANG-Jerry完成签到,获得积分10
3秒前
霸气南珍发布了新的文献求助10
4秒前
4秒前
4秒前
Giroro_roro发布了新的文献求助10
4秒前
lcj2022发布了新的文献求助10
5秒前
Lizhiiiy发布了新的文献求助10
6秒前
6秒前
优雅采文发布了新的文献求助10
6秒前
还好完成签到 ,获得积分10
7秒前
rtx00发布了新的文献求助10
8秒前
8秒前
star完成签到 ,获得积分10
8秒前
cslghe完成签到,获得积分10
8秒前
10秒前
10秒前
10秒前
Mingtiaoxiyue发布了新的文献求助20
10秒前
muxiangrong完成签到,获得积分0
10秒前
高万发布了新的文献求助10
11秒前
11秒前
11秒前
lcj2022完成签到,获得积分20
11秒前
张成完成签到,获得积分10
11秒前
12秒前
jkr完成签到,获得积分10
12秒前
syh发布了新的文献求助10
13秒前
希望天下0贩的0应助十五采纳,获得10
14秒前
14秒前
14秒前
15秒前
15秒前
15秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Picture Books with Same-sex Parented Families: Unintentional Censorship 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
A Preliminary Study on Correlation Between Independent Components of Facial Thermal Images and Subjective Assessment of Chronic Stress 500
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 360
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3970949
求助须知:如何正确求助?哪些是违规求助? 3515634
关于积分的说明 11179061
捐赠科研通 3250769
什么是DOI,文献DOI怎么找? 1795474
邀请新用户注册赠送积分活动 875831
科研通“疑难数据库(出版商)”最低求助积分说明 805188