15,530 Hits in 6.0 sec

Compressed Hierarchical Representations for Multi-Task Learning and Task Clustering [article]

João Machado de Freitas, Sebastian Berg, Bernhard C. Geiger, Manfred Mücke
2022 arXiv   pre-print
In this paper, we frame homogeneous-feature multi-task learning (MTL) as a hierarchical representation learning problem, with one task-agnostic and multiple task-specific latent representations.  ...  Drawing inspiration from the information bottleneck principle and assuming an additive independent noise model between the task-agnostic and task-specific latent representations, we limit the information  ...  The COMET -Competence Centers for Excellent Technologies program is supported by the Austrian Federal Ministries for Climate Action, Environment, Energy, Mobility, Innovation and Technology (BMK) and for  ... 
arXiv:2205.15882v1 fatcat:cq4t7z5pxjgbnmfkeh4vpqhceu

Multi-task representations in human cortex transform along a sensory-to-motor hierarchy [article]

Takuya Ito, John D Murray
2021 bioRxiv   pre-print
We found a cortical topography of representational alignment following a hierarchical sensory-association-motor gradient, revealing compression-then-expansion of multi-task dimensionality along this gradient  ...  Compression-then-expansion organization in models emerged exclusively in a training regime where internal representations are highly optimized for sensory-to-motor transformation, and not under generic  ...  The authors acknowledge the Yale Center for Research Computing at Yale University for providing access to the Grace cluster and associated research computing resources.  ... 
doi:10.1101/2021.11.29.470432 fatcat:45xwjxa2gbfftj5mos44p2q7xa

Hierarchical Self-Supervised Learning for Medical Image Segmentation Based on Multi-Domain Data Aggregation [article]

Hao Zheng, Jun Han, Hongxiao Wang, Lin Yang, Zhuo Zhao, Chaoli Wang, Danny Z. Chen
2021 arXiv   pre-print
In many scenarios, unannotated images are abundant and easy to acquire. Self-supervised learning (SSL) has shown great potentials in exploiting raw data information and representation learning.  ...  Our extensive experiments show that multi-domain joint pre-training benefits downstream segmentation tasks and outperforms single-domain pre-training significantly.  ...  National Science Foundation through grants IIS-1455886, CCF-1617735, CNS-1629914, and IIS-1955395.  ... 
arXiv:2107.04886v1 fatcat:rzi2htp4izgezaxb3i43hngk2m

Multi-document Summarization via Deep Learning Techniques: A Survey

Congbo Ma, Wei Emma Zhang, Mingyu Guo, Hu Wang, QUAN Z. Sheng
2022 ACM Computing Surveys  
Multi-document summarization (MDS) is an effective tool for information aggregation that generates an informative and concise summary from a cluster of topic-related documents.  ...  Our survey, the first of its kind, systematically overviews the recent deep learning based MDS models.  ...  • We propose a categorization scheme to organize current research and provide a comprehensive review for deep learning based MDS techniques, including deep learning based models, objective functions, benchmark  ... 
doi:10.1145/3529754 fatcat:r4lngnzrgjbfziazokpd2c5s44

Multi-document Summarization via Deep Learning Techniques: A Survey [article]

Congbo Ma, Wei Emma Zhang, Mingyu Guo, Hu Wang, Quan Z. Sheng
2021 arXiv   pre-print
Multi-document summarization (MDS) is an effective tool for information aggregation that generates an informative and concise summary from a cluster of topic-related documents.  ...  Our survey, the first of its kind, systematically overviews the recent deep learning based MDS models.  ...  Multi-modality for Multi-document Summarization Multi-modal learning has led to successes in many deep learning tasks, such as Visual Language Navigation [133] and Visual Question Answering [6] .  ... 
arXiv:2011.04843v3 fatcat:zfi52xxef5g2tjkaw6hgjpwa5i

Multi-task Clustering of Human Actions by Sharing Information

Xiaoqiang Yan, Shizhe Hu, Yangdong Ye
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
However, using shared information to improve multi-task human action clustering has never been considered before, and cannot be achieved using existing clustering methods.  ...  In this work, we present a novel and effective Multi-Task Information Bottleneck (MTIB) clustering method, which is capable of exploring the shared information between multiple action clustering tasks  ...  So the goal of our multi-task clustering method is to learn a good compressed representation p(t k |x k ) of X k to T k from its own feature variable Y k .  ... 
doi:10.1109/cvpr.2017.431 dblp:conf/cvpr/YanHY17 fatcat:sp6yaej7w5bsxprun6n2al2zoq

User Multi-Interest Modeling for Behavioral Cognition [article]

Bei Yang, Ke Liu, Xiaoxiao Xu, Renjun Xu, Qinghui Sun, Hong Liu, Huan Xu
2022 arXiv   pre-print
Experiments on several benchmark datasets show that our approach works well and outperforms state-of-the-art unsupervised representation methods in different downstream tasks.  ...  With the help of a novel attention module which can learn multi-interests of user, the second sub-module achieves almost lossless dimensionality reduction.  ...  They built a personalized memorization for each user, which remembers both intrinsic user tastes and multi-facet user interests with the learned while compressed memory. Pi et al.  ... 
arXiv:2110.11337v3 fatcat:siuxbxijvbaylmcoyb5nyqada4

Multi-task compressive sensing with Dirichlet process priors

Yuting Qi, Dehong Liu, David Dunson, Lawrence Carin
2008 Proceedings of the 25th international conference on Machine learning - ICML '08  
the appropriate sharing mechanisms as well as CS inversion for each task.  ...  Compressive sensing (CS) is an emerging field that, under appropriate conditions, can significantly reduce the number of measurements required for a given signal.  ...  Figure 1 .Figure 2 . 12 Ten template signals for 10-cluster case. Multi-task CS inversion error (%) for DP-MT and MT * CS for the ten-cluster case.  ... 
doi:10.1145/1390156.1390253 dblp:conf/icml/QiLDC08 fatcat:tv4r34wva5ghjlojo5gpgzpbxm

Structural Entropy Guided Graph Hierarchical Pooling [article]

Junran Wu, Xueyuan Chen, Ke Xu, Shangzhe Li
2022 arXiv   pre-print
Specifically, without assigning the layer-specific compression quota, a global optimization algorithm is designed to generate the cluster assignment matrices for pooling at once.  ...  However, because of the fixed compression quota and stepwise pooling design, these hierarchical pooling methods still suffer from local structure damage and suboptimal problem.  ...  Note that we add an additional linear layer after each SEP or SEP-U layer to learn more task-specific node representations.  ... 
arXiv:2206.13510v1 fatcat:ejdfchnq2nga3csoi44qftov5u

A Meta-learning based Graph-Hierarchical Clustering Method for Single Cell RNA-Seq Data [article]

Zixiang Pan, Yuefan Lin, Haokun Zhang, Yuansong Zeng, Weijiang Yu, Yuedong Yang
2022 bioRxiv   pre-print
Here, we present MeHi-SCC, a method which utilized meta-learning protocol and brought in multi scRNA-seq datasets' information in order to assist graph-based hierarchical sub-clustering process.  ...  In result, MeHi-SCC outperformed current-prevailing scRNA clustering methods and successfully identified cell subtypes in two large scale cell atlas.  ...  ACKNOWLEDGMENT This work was sponsored by National Natural Science Foundation of China (61772566) and Guangdong Key Field R&D Plan (2019B020228001).  ... 
doi:10.1101/2022.09.06.506784 fatcat:zk5ebjxwsja2pib52mug63bvny

Accurate Learning of Graph Representations with Graph Multiset Pooling [article]

Jinheon Baek, Minki Kang, Sung Ju Hwang
2021 arXiv   pre-print
Moreover, our methods can be easily extended to the previous node clustering approaches for hierarchical graph pooling.  ...  Recently proposed hierarchical graph pooling methods, on the other hand, may yield the same representation for two different graphs that are distinguished by the Weisfeiler-Lehman test, as they suboptimally  ...  ., 2017) , we regard the graph representation learning as a multi-set encoding problem.  ... 
arXiv:2102.11533v4 fatcat:qdqktuojbbbx7ofm6dgqzmevx4

Hierarchical Learning Using Deep Optimum-Path Forest

Luis C.S. Afonso, Clayton R. Pereira, Silke A.T. Weber, Christian Hook, Alexandre X. Falcão, João P. Papa
2020 Journal of Visual Communication and Image Representation  
In this work, we are interested in developing tools for the automatic identification of Parkinson's disease using machine learning and the concept of BoVW.  ...  The proposed approach concerns a hierarchical-based learning technique to design visual dictionaries through the Deep Optimum-Path Forest classifier.  ...  Acknowledgments The authors are grateful to FAPESP grants #2013/07375-0, #2014/12236-1 and #2016/19403-6, Capes, and CNPq grants #306166/2014-3, #307066/2017-7, and #303808/2018-7.  ... 
doi:10.1016/j.jvcir.2020.102823 fatcat:qpswddq7evechbgllmtqw7ot2u

Representation Similarity Analysis for Efficient Task Taxonomy & Transfer Learning

Kshitij Dwivedi, Gemma Roig
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We demonstrate the effectiveness and efficiency of our method for generating task taxonomy on Taskonomy dataset.  ...  We next evaluate the relationship of RSA with the transfer learning performance on Taskonomy tasks and a new task: Pascal VOC semantic segmentation.  ...  We thank Taskonomy authors for the support and the code.  ... 
doi:10.1109/cvpr.2019.01267 dblp:conf/cvpr/DwivediR19 fatcat:gzibhhjgtfhcnjv2iutrtyrlpy

MethylNet: A Modular Deep Learning Approach to Methylation Prediction [article]

Joshua J Levy, Alexander J Titus, Curtis L Petersen, David Chen, Lucas A Salas, Brock C Christensen
2019 bioRxiv   pre-print
We interrogate the learned features from a pan-cancer classification to show high fidelity clustering of cancer subtypes, and compare the importance assigned to CpGs for the age and cell-type analyses  ...  However, a generalized and user-friendly approach for execution, training, and interpreting deep learning models for methylation data is lacking.  ...  Fine-tune the model's prediction and feature extraction layers end-to-end for the tasks of multi-output regression and classification tasks.  ... 
doi:10.1101/692665 fatcat:gj6z2go2dbabvn3aqsz6ecbbm4

MxPool: Multiplex Pooling for Hierarchical Graph Representation Learning [article]

Yanyan Liang, Yanfeng Zhang, Dechao Gao, Qian Xu
2020 arXiv   pre-print
In this paper, we propose MxPool, which concurrently uses multiple graph convolution/pooling networks to build a hierarchical learning structure for graph representation learning tasks.  ...  diameter, and clustering coefficient).  ...  In this paper, we propose MxPool in hierarchical graph representation learning for graph classification tasks 2 .  ... 
arXiv:2004.06846v1 fatcat:ymocwmhknfamddcy6sxauwmwne
« Previous Showing results 1 — 15 out of 15,530 results