Filters








4,324 Hits in 3.8 sec

A probabilistic constrained clustering for transfer learning and image category discovery [article]

Yen-Chang Hsu, Zhaoyang Lv, Joel Schlosser, Phillip Odom, Zsolt Kira
2018 arXiv   pre-print
Neural network-based clustering has recently gained popularity, and in particular a constrained clustering formulation has been proposed to perform transfer learning and image category discovery using  ...  performance on both supervised learning and unsupervised transfer learning.  ...  Setup for image category discovery We follow the unsupervised transfer learning procedures described in [11] to evaluate the clustering performance with noisy constraints.  ... 
arXiv:1806.11078v1 fatcat:nk7kvto33fao7dsp6vvzjlr2pa

Discovering Latent Domains for Multisource Domain Adaptation [chapter]

Judy Hoffman, Brian Kulis, Trevor Darrell, Kate Saenko
2012 Lecture Notes in Computer Science  
Our discovery method is based on a novel hierarchical clustering technique that uses available object category information to constrain the set of feasible domain separations.  ...  In this paper, we present both a novel domain transform mixture model which outperforms a single transform model when multiple domains are present, and a novel constrained clustering method that successfully  ...  Our domain discovery method is based on a constrained clustering method, a topic of active research for several years. The most related work to our approach is based on constrained k-means.  ... 
doi:10.1007/978-3-642-33709-3_50 fatcat:afycjfq3czbudhazcc4odldup4

Towards Open-Set Object Detection and Discovery [article]

Jiyang Zheng, Weihao Li, Jie Hong, Lars Petersson, Nick Barnes
2022 arXiv   pre-print
With this method, a detector is able to detect objects belonging to known classes and define novel categories for objects of unknown classes with minimal supervision.  ...  In order to address this problem, we present a new task, namely Open-Set Object Detection and Discovery (OSODD).  ...  [19] formulated the task of novel class discovery (NCD), which clusters the unlabelled images into novel categories using deep transfer clustering.  ... 
arXiv:2204.05604v1 fatcat:adntvubdmjg3nfwirbvzy6idiu

DeepSym: Deep Symbol Generation and Rule Learning from Unsupervised Continuous Robot Interaction for Planning [article]

Alper Ahmetoglu, M. Yunus Seker, Justus Piater, Erhan Oztop, Emre Ugur
2021 arXiv   pre-print
In order to form action-grounded object, effect, and relational categories, we employ a binarized bottleneck layer of a predictive, deep encoder-decoder network that takes as input the image of the scene  ...  Towards this goal, we propose a novel and general method that finds action-grounded, discrete object and effect categories and builds probabilistic rules over them that can be used in complex action planning  ...  Acknowledgements This research was supported by TÜBİTAK (The Scientific and Technological Research Council of Turkey) ARDEB 1001 program (project number: 120E274).  ... 
arXiv:2012.02532v2 fatcat:wk6ia4eexvhofcilrb5zntoqh4

Author Index

2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
Context-Aware Category Discovery Reading Between The Lines: Object Localization Using Implicit Cues from Image Tags Far-Sighted Active Learning on a Budget for Image and Video Recognition Collect-Cut  ...  Connectivity Constraints for Reconstruction of 3D Line Segments from Images Tommasi, Tatiana Safety in Numbers: Learning Categories from Few Examples with Multi Model Knowledge Transfer Tong, Yan Workshop  ... 
doi:10.1109/cvpr.2010.5539913 fatcat:y6m5knstrzfyfin6jzusc42p54

Research on Artificial Intelligence Frontier Recognition Based on LDA

Ting Xie, Ping Qin, Juehu Yan
2018 OALib  
It is of great significance for the state, institutions and researchers to grasp the research frontier in a timely and accurate manner.  ...  and clustering research.  ...  very detailed manner for easy understanding and analysis. sourc, transfer, music, target, spatial, data, learn, risk, system, discoveri Chaotic music generation system using music conductor gesture Open  ... 
doi:10.4236/oalib.1105005 fatcat:rubbdo2n5jec5hi4wxee2urfju

Weakly Supervised Object Co-Localization via Sharing Parts Based on a Joint Bayesian Model

Lu Wu, Quan Liu
2018 Symmetry  
Objects in images are characterized by intra-class variation, inter-class diversity, and noisy images. These characteristics pose a challenge to object localization.  ...  object, parts and features within and between-class; (2) Labels are given at class level to provide strong supervision for features and corresponding parts; (3) Noisy images are considered by leveraging  ...  Acknowledgments: The authors would like to acknowledge funding from the National Science Foundation Committee (NSFC) of China (Grant No. 51675389 and No. 51475347) and the contributions of all collaborators  ... 
doi:10.3390/sym10050142 fatcat:uco5irnwnbb7hh5f3g5zqt4zcm

Superpixel-Based Conditional Random Fields (SuperCRF): Incorporating Global and Local Context for Enhanced Deep Learning in Melanoma Histopathology

Konstantinos Zormpas-Petridis, Henrik Failmezger, Shan E Ahmed Raza, Ioannis Roxanis, Yann Jamin, Yinyin Yuan
2019 Frontiers in Oncology  
We develop SuperCRF by training a state-of-art deep learning spatially constrained- convolution neural network (SC-CNN) to detect and classify cells from 105 high-resolution (20×) H&E-stained slides of  ...  ) given by a superpixel-based machine learning framework.  ...  Our proposed framework connects deep learning and classical image processing using probabilistic graphical models.  ... 
doi:10.3389/fonc.2019.01045 pmid:31681583 pmcid:PMC6798642 fatcat:a25nszx7dbby5ccei5eiuerjcq

Optimization Methods for Evaluating PEV Charging Considering Customer Behavior

Di Wu, Xinda Ke, Nikitha Radhakrishnan, Andrew Reiman
2018 2018 IEEE Power & Energy Society General Meeting (PESGM)  
Objects in images are characterized by intra-class variation, inter-class diversity, and noisy images. These characteristics pose a challenge to object localization.  ...  object, parts and features within and between-class; (2) Labels are given at class level to provide strong supervision for features and corresponding parts; (3) Noisy images are considered by leveraging  ...  Acknowledgments: The authors would like to acknowledge funding from the National Science Foundation Committee (NSFC) of China (Grant No. 51675389 and No. 51475347) and the contributions of all collaborators  ... 
doi:10.1109/pesgm.2018.8586401 fatcat:fk2dywywpfhipegjwwaoqe2nl4

How to Grow a Mind: Statistics, Structure, and Abstraction

J. B. Tenenbaum, C. Kemp, T. L. Griffiths, N. D. Goodman
2011 Science  
This review describes recent approaches to reverse-engineering human learning and cognitive development and, in parallel, engineering more humanlike machine learning systems.  ...  : How does abstract knowledge guide learning and reasoning from sparse data?  ...  Conventional algorithms for unsupervised structure discovery in statistics and machine learninghierarchical clustering, principal components analysis, multidimensional scaling, clique detectionassume a  ... 
doi:10.1126/science.1192788 pmid:21393536 fatcat:sh4diud5l5g6hkdw7u2eptlpou

Unsupervised Part Discovery from Contrastive Reconstruction [article]

Subhabrata Choudhury, Iro Laina, Christian Rupprecht, Andrea Vedaldi
2022 arXiv   pre-print
The goal of self-supervised visual representation learning is to learn strong, transferable image representations, with the majority of research focusing on object or scene level.  ...  First, we construct a proxy task through a set of objectives that encourages the model to learn a meaningful decomposition of the image into its parts.  ...  Acknowledgements and Funding Disclosure S. C. is supported by a scholarship sponsored by Facebook. I.  ... 
arXiv:2111.06349v2 fatcat:qxtuzama7vfvdmkg7gthkaeeh4

2019 Index IEEE Transactions on Knowledge and Data Engineering Vol. 31

2020 IEEE Transactions on Knowledge and Data Engineering  
Augusto, A., +, TKDE April 2019 686-705 Latent Ability Model: A Generative Probabilistic Learning Framework for Workforce Analytics.  ...  ., +, TKDE Feb. 2019 201-213 A General Domain Specific Feature Transfer Framework for Hybrid Domain Adaptation.  ... 
doi:10.1109/tkde.2019.2953412 fatcat:jkmpnsjcf5a3bhhf4ian66mj5y

Unsupervised Learning of Category-Specific Symmetric 3D Keypoints from Point Sets [article]

Clara Fernandez-Labrador, Ajad Chhatkuli, Danda Pani Paudel, Jose J. Guerrero, Cédric Demonceaux, Luc Van Gool
2021 arXiv   pre-print
Automatic discovery of category-specific 3D keypoints from a collection of objects of some category is a challenging problem.  ...  The usage of symmetry prior leads us to learn stable keypoints suitable for higher misalignments.  ...  Automatic discovery of category-specific 3D keypoints from a collection of objects of some category is a challenging problem.  ... 
arXiv:2003.07619v3 fatcat:sukmkgxk4jeynh6a4zhtjp7wxe

Machine Learning with World Knowledge: The Position and Survey [article]

Yangqiu Song, Dan Roth
2017 arXiv   pre-print
Machine learning has become pervasive in multiple domains, impacting a wide variety of applications, such as knowledge discovery and data mining, natural language processing, information retrieval, computer  ...  Two essential problems of machine learning are how to generate features and how to acquire labels for machines to learn.  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon.  ... 
arXiv:1705.02908v1 fatcat:t4fypa6h3vampcp64eosvppsfe

Retinal Image Classification by Self-supervised Fuzzy Clustering Network

Yueguo Luo, Jing Pan, Shaoshuai Fan, Zeyu Du, Guanghua Zhang
2020 IEEE Access  
Specifically, we propose a Self-supervised Fuzzy Clustering Network (SFCN) by a feature learning module, reconstruction module, and a fuzzy self-supervision module.  ...  INDEX TERMS Retinal image classification, self-supervised, fuzzy clustering, unsupervised learning. This work is licensed under a Creative Commons Attribution 4.0 License.  ...  [9] proposed a robust hybrid probabilistic learning approach that appropriately combines the advantages both of the generative and discriminative model for challenging problem in retinal image classification  ... 
doi:10.1109/access.2020.2994047 fatcat:t77mpusgerb5tb2g7vuitxdseq
« Previous Showing results 1 — 15 out of 4,324 results