paper-with-me

홈 › Papers

Information-theoretical label embeddings for large-scale image classification

2016-07-19 · François Chollet

We present a method for training multi-label, massively multi-class image classification models, that is faster and more accurate than supervision via a sigmoid cross-entropy loss (logistic regression). Our method consists in embedding high-dimensional sparse labels onto a lower-dimensional dense sphere of unit-normed vectors, and treating the classification problem as a cosine proximity regression problem on this sphere. We test our method on a dataset of 300 million high-resolution images with 17,000 labels, where it yields considerably faster convergence, as well as a 7% higher mean average precision compared to logistic regression.

📄 PDF Abstract BibTeX arXiv:1607.05691

Code (0)

등록된 구현이 없습니다.

Tasks

ClassificationGeneral Classificationimage-classificationImage Classificationregression

Similar Papers 제목 키워드 기반

Breaking Secure Aggregation: Label Leakage from Aggregated Gradients in Federated Learning

2024-06-22 · Zhibo Wang, Zhiwei Chang, Jiahui Hu, Xiaoyi Pang 외

Federated Learning (FL) exhibits privacy vulnerabilities under gradient inversion attacks (GIAs), which can extract private information from individual gradients. To enhance privacy, FL incorporates Secure Aggregation (S…

Federated LearningInference Attack

Modeling Label Space Interactions in Multi-label Classification using Box Embeddings

2021-09-29 · ICLR 2022 · Anonymous

Multi-label classification is a challenging structured prediction task in which a set of output class labels are predicted for each input. Real-world datasets often have natural or latent taxonomic relationships between …

ClassificationHierarchical Multi-label ClassificationInductive BiasMulti-Label Classification+2

Semi-supervised Clustering Through Representation Learning of Large-scale EHR Data

2025-05-27 · Linshanshan Wang, Mengyan Li, Zongqi Xia, Molei Liu 외

Electronic Health Records (EHR) offer rich real-world data for personalized medicine, providing insights into disease progression, treatment responses, and patient outcomes. However, their sparsity, heterogeneity, and hi…

Representation Learning

Leveraging Distributional Semantics for Multi-Label Learning

2017-09-18 · Rahul Wadbude, Vivek Gupta, Piyush Rai, Nagarajan Natarajan 외

We present a novel and scalable label embedding framework for large-scale multi-label learning a.k.a ExMLDS (Extreme Multi-Label Learning using Distributional Semantics). Our approach draws inspiration from ideas rooted …

Document EmbeddingMissing LabelsMulti-Label LearningWord Embeddings

Fast Label Embeddings for Extremely Large Output Spaces

2015-03-30 · Paul Mineiro, Nikos Karampatziakis

Many modern multiclass and multilabel problems are characterized by increasingly large output spaces. For these problems, label embeddings have been shown to be a useful primitive that can improve computational and stati…