Filters








62,711 Hits in 7.9 sec

Multi-label Classification with Output Kernels [chapter]

Yuhong Guo, Dale Schuurmans
2013 Lecture Notes in Computer Science  
Although multi-label classification has become an increasingly important problem in machine learning, current approaches remain restricted to learning in the original label space (or in a simple linear  ...  Instead, we propose to use kernels on output label vectors to significantly expand the forms of label dependence that can be captured.  ...  Although kernel methods have been widely used for expanding input representations, kernels have yet to be used to explicitly capture nonlinear output structure in multi-label classification.  ... 
doi:10.1007/978-3-642-40991-2_27 fatcat:e5hn3xugljcufmhqyesa2ydxcm

A General Two-stage Multi-label Ranking Framework

Yanbing Xue, Milos Hauskrecht
2021 Proceedings of the ... International Florida Artificial Intelligence Research Society Conference  
In this paper we develop and study solutions for the multi-label ranking (MLR) problem.  ...  Briefly, the goal of multi-label ranking is not only to assign a set of relevant labels to a data instance but also to rank the labels according to their importance.  ...  (2) rank the relevant labels from the output of the existing multi-label classifier.  ... 
doi:10.32473/flairs.v34i1.128505 fatcat:llnsywnuhjb7xehv76naqh4cvu

A Survey on Multi-output Learning [article]

Donna Xu, Yaxin Shi, Ivor W. Tsang, Yew-Soon Ong, Chen Gong, Xiaobo Shen
2019 arXiv   pre-print
Inspired by big data, the 4Vs characteristics of multi-output imposes a set of challenges to multi-output learning, in terms of the volume, velocity, variety and veracity of the outputs.  ...  Then we present the paradigm on multi-output learning, including its myriads of output structures, definitions of its different sub-problems, model evaluation metrics and popular data repositories used  ...  Embedding methods can be used to compress a space by projecting the original space onto a lowerdimensional space, with the expected information preserved, such as label correlations and neighborhood structure  ... 
arXiv:1901.00248v2 fatcat:obc5iphzv5gj7ewut6lcqchqiy

A snapshot on nonstandard supervised learning problems: taxonomy, relationships, problem transformations and algorithm adaptations

David Charte, Francisco Charte, Salvador García, Francisco Herrera
2018 Progress in Artificial Intelligence  
This field is subdivided into multiple areas, among which the best known are supervised learning (e.g. classification and regression) and unsupervised learning (e.g. clustering and association rules).  ...  Within supervised learning, most studies and research are focused on well known standard tasks, such as binary classification, multiclass classification and regression with one dependent variable.  ...  Label ranking In a label ranking (LR) problem [57, 113] the objective is not to find a function able to choose one or several labels from the label space.  ... 
doi:10.1007/s13748-018-00167-7 fatcat:lqg23f33hbfu3heht2swg7k2hy

Multi-label Learning Based on Kernel Extreme Learning Machine

Fangfang Luo, Wenzhong Guo, Fangwan Huang, Guolong Chen
2018 DEStech Transactions on Computer Science and Engineering  
Due to the huge solution space, the problem becomes more complex. Therefore, we propose a multi-label algorithm based on kernel learning machine in this paper.  ...  In recent years, with the increase of data scale, multi-label learning with large scale class labels has turned out to be the research hotspots.  ...  ALGORITHM MODEL Multi The existing ELM structure [1] for solving multi-label problem is shown in Figure 1 layer output matrix H and hidden layer output weight  , which can be presented by Eq.  ... 
doi:10.12783/dtcse/csae2017/17476 fatcat:362vdxzye5bcrbq2ddw5nnab2i

The Emerging Trends of Multi-Label Learning [article]

Weiwei Liu, Xiaobo Shen, Haobo Wang, Ivor W. Tsang
2020 arXiv   pre-print
Exabytes of data are generated daily by humans, leading to the growing need for new efforts in dealing with the grand challenges for multi-label learning brought by big data.  ...  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  ...  Low-Rank and Embedding Methods As discussed in §2.1, the existence of label correlations usually implies the output space is low-rank.  ... 
arXiv:2011.11197v2 fatcat:hu6w4vgnwbcqrinrdfytmmjbjm

ReliefF for Hierarchical Multi-label Classification [chapter]

Ivica Slavkov, Jana Karcheska, Dragi Kocev, Slobodan Kalajdziski, Sašo Džeroski
2014 Lecture Notes in Computer Science  
More specifically, we adapt the RReliefF algorithm for regression, for the task of hierarchical multi-label classification (HMC).  ...  Our work in this paper, focuses on extending a feature ranking algorithm that can be used as a filter method for a specific type of structured data.  ...  We would like to acknowledge the support of the European Commission through the project MAESTRA -Learning from Massive, Incompletely annotated, and Structured Data (Grant number ICT-2013-612944).  ... 
doi:10.1007/978-3-319-08407-7_10 fatcat:s7oziycdt5f6xifkxytq5vqvxm

HMC-ReliefF: Feature ranking for hierarchical multi-label classification

Ivica Slavkov, Jana Karcheska, Dragi Kocev, Saso Dzeroski
2018 Computer Science and Information Systems  
Our work in this paper focuses on the development and analysis of the HMC-ReliefF algorithm, which is a feature relevance (ranking) algorithm for the task of Hierarchical Multi-label Classification (HMC  ...  The basis of the algorithm is the RReliefF algorithm for regression that is adapted for hierarchical multi-label target variables.  ...  ) multi-label classification.  ... 
doi:10.2298/csis170115043s fatcat:bhwxk4uzzbfstp6nreqwy2tznq

An Efficient Stacking Model of Multi-label Classification Based on Pareto Optimum

Wei Weng, Chin-Ling Chen, Shun-Xiang Wu, Yu-Wen Li, Juan Wen
2019 IEEE Access  
Among those algorithms binary relevance (BR) is a widely used framework for multi-label classification. It constructs binary classifiers for each label by means of one-vs-rest style.  ...  The final binary classifiers for each label are constructed based on their corresponding reconstructed feature space.  ...  Inspired by the method for extracting label specific features introduced in ParetoFS [30] , we map the output space of base level to a multi-dimensional space with the help of ReliefF algorithm [31]  ... 
doi:10.1109/access.2019.2931451 fatcat:zwr5t5gxnfdd7pteaq74b4mqae

GUDN A novel guide network for extreme multi-label text classification [article]

Qing Wang, Hongji Shu, Jia Zhu
2022 arXiv   pre-print
The problem of extreme multi-label text classification (XMTC) is to recall some most relevant labels for a text from an extremely large label set.  ...  Also, we use the raw label semantics to effectively explore the latent space between texts and labels, which can further improve predicted accuracy.  ...  ., 2017] explored the latent space between texts and labels by using a sparse linear network making new development for the multi-label classification tasks.  ... 
arXiv:2201.11582v1 fatcat:vz3vxkwdi5hudikkz2pzqtvdda

Deep tree-ensembles for multi-output prediction [article]

Felipe Kenji Nakano, Konstantinos Pliakos, Celine Vens
2021 arXiv   pre-print
In this paper, we specifically focus on two structured output prediction tasks, namely multi-label classification and multi-target regression.  ...  Despite that, these approaches simply employ label classification probabilities as induced features and primarily focus on traditional classification and regression tasks, leaving multi-output prediction  ...  Acknowledgements The authors acknowledge the support from the Research Fund Flanders (through research project G080118N) and from the Flemish Government (AI Research Program).  ... 
arXiv:2011.02829v2 fatcat:44jd5rm6kbcdhpkbqbm5isfl2q

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  
embedding space; 2) the ranking loss not only improves the training efficiency and accuracy but also can be extended to handle noisy labeled data; 3) the efficient attention mechanism improves feature  ...  The proposed method has three major advantages: 1) the autoencoder simultaneously considers the inter-label dependencies and the feature-label dependencies, by projecting labels and features onto a common  ...  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
embedding space; 2) the ranking loss not only improves the training efficiency and accuracy but also can be extended to handle noisy labeled data; 3) the efficient attention mechanism improves feature  ...  The proposed method has three major advantages: 1) the autoencoder simultaneously considers the inter-label dependencies and the feature-label dependencies, by projecting labels and features onto a common  ...  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

Preference Neural Network [article]

Ayman Elgharabawy, Mukesh Prasad, Chin-Teng Lin
2021 arXiv   pre-print
PNN inputs represent data features and output neurons represent label indexes.  ...  PNN also solves the Multi-label ranking problem, where labels may have indifference preference orders or subgroups are equally ranked.  ...  ACKNOWLEDGEMENT This work was supported in part by the Australian Research Council (ARC) under discovery grant DP180100670 and DP180100656.  ... 
arXiv:1904.02345v3 fatcat:73ftmvi5nbfyxn7oon7zq7k2te

Log-time and Log-space Extreme Classification [article]

Kalina Jasinska, Nikos Karampatziakis
2016 arXiv   pre-print
We present LTLS, a technique for multiclass and multilabel prediction that can perform training and inference in logarithmic time and space.  ...  LTLS embeds large classification problems into simple structured prediction problems and relies on efficient dynamic programming algorithms for inference.  ...  Introduction Extreme multi-class and multi-label classification refers to problems where the size C of the output space is extremely large.  ... 
arXiv:1611.01964v1 fatcat:il4vnwmztrbgfdl2iolsink3nu
« Previous Showing results 1 — 15 out of 62,711 results