Filters








50,502 Hits in 5.4 sec

Locally Non-linear Embeddings for Extreme Multi-label Learning [article]

Kush Bhatia and Himanshu Jain and Purushottam Kar and Prateek Jain and Manik Varma
2015 arXiv   pre-print
The objective in extreme multi-label learning is to train a classifier that can automatically tag a novel data point with the most relevant subset of labels from an extremely large label set.  ...  The main technical contribution in X-One is a formulation for learning a small ensemble of local distance preserving embeddings which can accurately predict infrequently occurring (tail) labels.  ...  Recently, tree based methods [1, 15, 2] have also become popular for extreme multi-label learning as they enjoy significant accuracy gains over the existing embedding methods.  ... 
arXiv:1507.02743v1 fatcat:akyvda6lqzhn3kufwyosfdrn5u

Deep Extreme Multi-label Learning [article]

Wenjie Zhang, Junchi Yan, Xiangfeng Wang, Hongyuan Zha
2018 arXiv   pre-print
In this paper, we propose a practical deep embedding method for extreme multi-label classification, which harvests the ideas of non-linear embedding and graph priors-based label space modeling simultaneously  ...  Extreme multi-label learning (XML) or classification has been a practical and important problem since the boom of big data.  ...  We also would like to thank Liwei Wang for discussions on label graph modeling.  ... 
arXiv:1704.03718v4 fatcat:iybndvgxv5cwre5yg6ruakl32y

Sparse Local Embeddings for Extreme Multi-label Classification

Kush Bhatia, Himanshu Jain, Purushottam Kar, Manik Varma, Prateek Jain
2015 Neural Information Processing Systems  
The objective in extreme multi-label learning is to train a classifier that can automatically tag a novel data point with the most relevant subset of labels from an extremely large label set.  ...  The main technical contribution in SLEEC is a formulation for learning a small ensemble of local distance preserving embeddings which can accurately predict infrequently occurring (tail) labels.  ...  Acknowledgments We are grateful to Abhishek Kadian for helping with the experiments. Himanshu Jain is supported by a Google India PhD Fellowship at IIT Delhi  ... 
dblp:conf/nips/BhatiaJKVJ15 fatcat:y42jjydpzzhcjkkrmqysys7pf4

DiSMEC - Distributed Sparse Machines for Extreme Multi-label Classification [article]

Rohit Babbar, Bernhard Shoelkopf
2016 arXiv   pre-print
Most state-of-the-art approaches for extreme multi-label classification attempt to capture correlation among labels by embedding the label matrix to a low-dimensional linear sub-space.  ...  Extreme multi-label classification refers to supervised multi-label learning involving hundreds of thousands or even millions of labels.  ...  Label matrix compression is then achieved by learning non-linear embeddings instead of linear mappings as in previous approaches.  ... 
arXiv:1609.02521v1 fatcat:tol2zxjnxnbvdgzasmvi5p7f3i

Breaking the Glass Ceiling for Embedding-Based Classifiers for Large Output Spaces

Chuan Guo, Ali Mousavi, Xiang Wu, Daniel Niels Holtmann-Rice, Satyen Kale, Sashank J. Reddi, Sanjiv Kumar
2019 Neural Information Processing Systems  
In extreme classification settings, embedding-based neural network models are currently not competitive with sparse linear and tree-based methods in terms of accuracy.  ...  To this end, we propose GLaS, a new regularizer for embedding-based neural network approaches.  ...  extreme multi-label classification problems.  ... 
dblp:conf/nips/GuoMWHKRK19 fatcat:vi55xnimrbao5h3kpvvxdkb4ua

The Emerging Trends of Multi-Label Learning [article]

Weiwei Liu, Xiaobo Shen, Haobo Wang, Ivor W. Tsang
2020 arXiv   pre-print
For example, extreme multi-label classification is an active and rapidly growing research area that deals with classification tasks with an extremely large number of classes or labels; utilizing massive  ...  Besides these, there are tremendous efforts on how to harvest the strong learning capability of deep learning to better capture the label dependencies in multi-label learning, which is the key for deep  ...  Word embeddings [63] have been successfully used for learning non-linear representations of text data for natural language processing (NLP) tasks, such as understanding word and document semantics and  ... 
arXiv:2011.11197v2 fatcat:hu6w4vgnwbcqrinrdfytmmjbjm

Ranking-Based Autoencoder for Extreme Multi-label Classification

Bingyu Wang, Li Chen, Wei Sun, Kechen Qin, Kefeng Li, Hui Zhou
2019 Proceedings of the 2019 Conference of the North  
Extreme Multi-label classification (XML) is an important yet challenging machine learning task, that assigns to each instance its most relevant candidate labels from an extremely large label collection  ...  The extremely large label collections yield challenges such as computational complexity, inter-label dependency and noisy labeling.  ...  We would also like to show our gratitude to our anonymous NAACL-HLT reviwers for the helpful suggestions to make the paper better.  ... 
doi:10.18653/v1/n19-1289 dblp:conf/naacl/WangCSQLZ19 fatcat:mgezaqg7nzeqjewqci57uduyxy

Ranking-Based Autoencoder for Extreme Multi-label Classification [article]

Bingyu Wang, Li Chen, Wei Sun, Kechen Qin, Kefeng Li, Hui Zhou
2019 arXiv   pre-print
Extreme Multi-label classification (XML) is an important yet challenging machine learning task, that assigns to each instance its most relevant candidate labels from an extremely large label collection  ...  The extremely large label collections yield challenges such as computational complexity, inter-label dependency and noisy labeling.  ...  We would also like to show our gratitude to our anonymous NAACL-HLT reviwers for the helpful suggestions to make the paper better.  ... 
arXiv:1904.05937v1 fatcat:dgpgy4wn4vezfmtaqhvgz6pwua

On Riemannian Approach for Constrained Optimization Model in Extreme Classification Problems [article]

Jayadev Naram, Tanmay Kumar Sinha, Pawan Kumar
2021 arXiv   pre-print
We propose a novel Riemannian method for solving the Extreme multi-label classification problem that exploits the geometric structure of the sparse low-dimensional local embedding models.  ...  The proposed approach is tested on several real world large scale multi-label datasets and its usefulness is demonstrated through numerical experiments.  ...  Locally Sparse Embedding for Extreme Classification Earlier embedding based methods assumed that the label matrix has low rank.  ... 
arXiv:2109.15021v1 fatcat:s2yv3dqivrferbgykmiev4eq7y

On-the-fly Global Embeddings Using Random Projections for Extreme Multi-label Classification [article]

Yashaswi Verma
2021 arXiv   pre-print
The goal of eXtreme Multi-label Learning (XML) is to automatically annotate a given data point with the most relevant subset of labels from an extremely large vocabulary of labels (e.g., a million labels  ...  Precisely, we investigate an on-the-fly global and structure preserving feature embedding technique using random projections whose learning phase is independent of training samples and label vocabulary  ...  INTRODUCTION E XTREME Multi-label Learning (or XML) is the problem of learning a classification model that can automatically assign a subset of the most relevant labels to a data point from an extremely  ... 
arXiv:1912.08140v2 fatcat:zgratigbczfj5kf4bww3dhhlha

A Novel Adaptive Multi-view Non-Negative Graph Semi-supervised ELM

Feng Zheng, Zeyu Liu, Yijian Chen, Jiacheng An, Yanyan Zhang
2020 IEEE Access  
matrix for multi-view and can not establish a non-liner model for unknown data.  ...  that is applicable to any single or multi-view data, and builds linear or non-linear models.  ...  First, the data is single view and the relationship between the data and the label is non-linear. Second, the data is multi-view and the relationship between the data and the label is non-linear.  ... 
doi:10.1109/access.2020.2998428 fatcat:5kuens7aibb35kgjkdyxovx76q

Multi-Label Patent Categorization with Non-Local Attention-Based Graph Convolutional Network

Pingjie Tang, Meng Jiang, Bryan (Ning) Xia, Jed W. Pitera, Jeffrey Welser, Nitesh V. Chawla
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
It employs a non-local attention mechanism to learn label representations in the same space of document representations for multi-label classification.  ...  It jointly learns the document-word associations and word-word co-occurrences to generate rich semantic embeddings of documents.  ...  Embedding based models. Embedding based methods employ compression functions to project label embedding to a lower dimensional linear subspace.  ... 
doi:10.1609/aaai.v34i05.6435 fatcat:z25s27ojf5f7vnbh7bgmjpdziu

Leveraging Distributional Semantics for Multi-Label Learning [article]

Rahul Wadbude, Vivek Gupta, Piyush Rai, Nagarajan Natarajan, Harish Karnick, Prateek Jain
2017 arXiv   pre-print
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 is novel in that it highlights interesting connections between label embedding methods used for multi-label learning and paragraph/document embedding methods commonly used for learning representations  ...  Unlike other embedding based methods, SLEEC has the ability to learn non-linear embeddings by aiming to preserve only local structures and example neighborhoods.  ... 
arXiv:1709.05976v3 fatcat:4idg7s27ivaadn443jponx5tfi

IIT BHU at TREC 2019 Incident Streams Track

Akanksha Mishra, Sukomal Pal
2019 Text Retrieval Conference  
We submitted two fully automatic runs for categorizing information within tweet into multiple highlevel information types and determining the criticality score for each tweet given in the test set.  ...  We used Sparse Local Embeddings for Extreme Multilabel Classification [1] for multi-label classification.  ...  Specifically, divide the dataset into several clusters, and in each cluster, it detects embedding vectors by capturing non-linear label correlation and preserving the pairwise distance between labels.  ... 
dblp:conf/trec/MishraP19 fatcat:b62xhuoiunfz3pfpmg6l4fjlj4

Intra-class multi-output regression based subspace analysis

S. Karthikeyan, Swapna Joshi, B.S. Manjunath, Scott Grafton
2012 2012 19th IEEE International Conference on Image Processing  
Our method leverages the multi-dimensional image labels that quantify the within class regression to learn the subspaces for recognition.  ...  To handle such scenarios, we formulate a supervised Non-negative Matrix Factorization (NMF) based subspace learning technique that simultaneously preserves the intra-class regression information (local  ...  Manifold learning algorithms [2] are geometrically motivated non-linear reduction methods; recently, [3] proposed a supervised manifold embedding extention.  ... 
doi:10.1109/icip.2012.6467074 dblp:conf/icip/KarthikeyanJMG12 fatcat:3df7d4uosbh7zhmwele5uvys7i
« Previous Showing results 1 — 15 out of 50,502 results