Filters








181,774 Hits in 6.1 sec

A risk minimization framework for domain adaptation

Bo Long, Sudarshan Lamkhede, Srinivas Vadrevu, Ya Zhang, Belle Tseng
2009 Proceeding of the 18th ACM conference on Information and knowledge management - CIKM '09  
A novel learning framework, Domain Transfer Risk Minimization (DTRM), is proposed based on this concept.  ...  DTRM simultaneously minimizes the empirical risk for the target and the regularized empirical risk for source domain.  ...  Based on the concept of the label-relation function, we propose a new framework, Domain Transfer Risk Minimization (DTRM), for domain adaptation.  ... 
doi:10.1145/1645953.1646123 dblp:conf/cikm/LongLVZT09 fatcat:vmpwsuag6jgjxcrxcy6dikirpe

Domain Adaptation via Maximizing Surrogate Mutual Information [article]

Haiteng Zhao, Chang Ma, Qinyu Chen, Zhi-Hong Deng
2022 arXiv   pre-print
In this work, we propose a novel framework called SIDA (Surrogate Mutual Information Maximization Domain Adaptation) with strong theoretical guarantees.  ...  In the framework, a surrogate joint distribution models the underlying joint distribution of the unlabeled target domain.  ...  In this work, we minimize the expected risk on the source domain and maximize MI, for minimizing the upper bound of expected risk on target domain.  ... 
arXiv:2110.12184v2 fatcat:3tgdw34uyfa2zoep55w2evyjhu

Generalization Bounds Derived IPM-Based Regularization for Domain Adaptation

Juan Meng, Guyu Hu, Dong Li, Yanyan Zhang, Zhisong Pan
2016 Computational Intelligence and Neuroscience  
In order to improve the generalization ability of domain adaption methods, we proposed a framework for domain adaptation combining source and target data, with a new regularizer which takes generalization  ...  With popular learning models, the empirical risk minimization is expressed as a general convex optimization problem and thus can be solved effectively by existing tools.  ...  However, the target samples are not enough to learn a predictor; that is, ≪ ; then domain adaptation minimize the convex combination of the source and the target empirical risk, for ∈ [0, 1), := ( ) +  ... 
doi:10.1155/2016/7046563 pmid:26819589 pmcid:PMC4707017 fatcat:6tmzzozp6zbavp4kocb37ui2bu

HierMUD: Hierarchical Multi-task Unsupervised Domain Adaptation between Bridges for Drive-by Damage Diagnosis [article]

Jingxiao Liu, Susu Xu, Mario Bergés, Hae Young Noh
2021 arXiv   pre-print
To this end, we introduce a new framework that transfers the model learned from one bridge to diagnose damage in another bridge without any labels from the target bridge.  ...  Our framework trains a hierarchical neural network model in an adversarial way to extract task-shared and task-specific features that are informative to multiple diagnostic tasks and invariant across multiple  ...  A generalization risk bound for unsupervised domain adaptation We first derive a new generalization risk bound for UDA by representing the original data distribution in a feature space, which has been  ... 
arXiv:2107.11435v1 fatcat:hhcgs3mm4bhp5a6snahfcyogmm

Patient safety risk factors in minimally invasive surgery: a validation study

Sharon P. Rodrigues, Moniek ter Kuile, Jenny Dankelman, Frank W. Jansen
2011 Gynecological Surgery  
This study was conducted to adapt and validate a patient safety (PS) framework for minimally invasive surgery (MIS) as a first step in understanding the clinical relevance of various PS risk factors in  ...  For seven of nine risk domains, Cronbach's alpha was sufficient (α>0.7).  ...  The authors alone are responsible for the content and writing of the paper. Gynecol Surg (2012) 9:265-270  ... 
doi:10.1007/s10397-011-0718-0 pmid:22837734 pmcid:PMC3401291 fatcat:xzylyxydgjdldfoynyyyrqbdla

Contradictory Structure Learning for Semi-supervised Domain Adaptation [article]

Can Qin, Lichen Wang, Qianqian Ma, Yu Yin, Huan Wang, Yun Fu
2021 arXiv   pre-print
To solve these challenges, we propose a novel framework for semi-supervised domain adaptation by unifying the learning of opposite structures (UODA).  ...  UODA consists of a generator and two classifiers (i.e., the source-scattering classifier and the target-clustering classifier), which are trained for contradictory purposes.  ...  The task losses of the two domains are summed for empirical risk minimization: Θ * G = arg min Θ G L src + L tar − βH src + λH tar . (9) The whole framework is trained in an end-to-end manner with the  ... 
arXiv:2002.02545v2 fatcat:xvyot7yg3nffterwvccmf2xyme

Domain adaptation of weighted majority votes via perturbed variation-based self-labeling [article]

Emilie Morvant
2014 arXiv   pre-print
In this work, we propose a framework to extend MinCq to a domain adaptation scenario.  ...  In machine learning, the domain adaptation problem arrives when the test (target) and the train (source) data are generated from different distributions.  ...  Discussion and future work We design a general PAC-Bayesian domain adaptation (DA) framework-PV-MinCq-for learning a target weighted majority vote over a set of real-valued functions.  ... 
arXiv:1410.0334v1 fatcat:ps75jotsdnejxexg6mgjsen4pe

Domain adaptation of weighted majority votes via perturbed variation-based self-labeling

Emilie Morvant
2015 Pattern Recognition Letters  
We tackle the PAC-Bayesian Domain Adaptation (DA) problem.  ...  Secondly, we propose an original process for tuning the hyperparameters. Our framework shows very promising results on a toy problem.  ...  Discussion and future work We design a general PAC-Bayesian domain adaptation (DA) framework-PV-MinCq-for learning a target weighted majority vote over a set of real-valued functions.  ... 
doi:10.1016/j.patrec.2014.08.013 fatcat:2uutvv2xofhm3jkapdk7g7pyre

An Overview of Transfer Learning and Computational CyberPsychology [chapter]

Zengda Guan, Tingshao Zhu
2013 Lecture Notes in Computer Science  
We finally give a transfer learning framework for Computational CyberPsychology, and describe how it can be implemented.  ...  We introduce Computational CyberPsychology at first, and then transfer learning, including sample selection bias and domain adaptation.  ...  [9] proposed a integrated optimization method for discriminative learning problem under covariant shift. Domain Adaptation To cope with domain adaptation, Blitzer et al.  ... 
doi:10.1007/978-3-642-37015-1_17 fatcat:cefyo552xbhvphwok2yx3dbfam

Active domain adaptation with noisy labels for multimedia analysis

Gaowen Liu, Yan Yan, Ramanathan Subramanian, Jingkuan Song, Guoyu Lu, Nicu Sebe
2015 World wide web (Bussum)  
Active learning (AL) and domain adaptation (DA) are two strategies to minimize the required amount of labeled data for model training.  ...  AL requires the domain expert to label a small number of highly informative examples to facilitate classification, while DA involves tuning the source domain knowledge for classification on the target  ...  Acknowledgments This work was partially supported by the MIUR Cluster project Active Ageing at Home, the EC project xLiMe and A*STAR Singapore under the Human-Centered Cyber-physical Systems (HCCS) grant  ... 
doi:10.1007/s11280-015-0343-3 fatcat:vh3wdd7nlvb47dkl7n47lrnlci

Entropy Minimization vs. Diversity Maximization for Domain Adaptation [article]

Xiaofu Wu, Suofei hang, Quan Zhou, Zhen Yang, Chunming Zhao, Longin Jan Latecki
2020 arXiv   pre-print
Entropy minimization has been widely used in unsupervised domain adaptation (UDA). However, existing works reveal that entropy minimization only may result into collapsed trivial solutions.  ...  In order to achieve the possible minimum target risk for UDA, we show that diversity maximization should be elaborately balanced with entropy minimization, the degree of which can be finely controlled  ...  CONCLUSION Entropy minimization has been shown to be a powerful tool for domain adaptation.  ... 
arXiv:2002.01690v1 fatcat:6ixceot25vfnph2fxvcs4jtigi

Transfer Adaptation Learning: A Decade Survey [article]

Lei Zhang, Xinbo Gao
2020 arXiv   pre-print
Conventional machine learning aims to find a model with the minimum expected risk on test data by minimizing the regularized empirical risk on the training data, which, however, supposes that the training  ...  Domain is referred to as the state of the world at a certain moment.  ...  ACKNOWLEDGMENT The author would like to thank the pioneer researchers in transfer learning, domain adaptation and other related fields. The author would also like to thank Dr. Mingsheng Long and Dr.  ... 
arXiv:1903.04687v2 fatcat:wurprqieffalnnp6isfkhh5y5i

Semi-supervised Domain Adaptation with Subspace Learning for visual recognition

Ting Yao, Yingwei Pan, Chong-Wah Ngo, Houqiang Li, Tao Mei
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
This paper proposes a novel domain adaptation framework, named Semi-supervised Domain Adaptation with Subspace Learning (SDASL), which jointly explores invariant lowdimensional structures across domains  ...  In many real-world applications, we are often facing the problem of cross domain learning, i.e., to borrow the labeled data or transfer the already learnt knowledge from a source domain to a target domain  ...  By consolidating the idea of semi-supervised learning and subspace learning for domain adaptation, this paper presents a novel Semi-supervised Domain Adaptation with Subspace Learning (SDASL) framework  ... 
doi:10.1109/cvpr.2015.7298826 dblp:conf/cvpr/YaoPNLM15 fatcat:uxgbcxd2kfagtj5nknmsmuo234

Learning Invariant Representation with Consistency and Diversity for Semi-supervised Source Hypothesis Transfer [article]

Xiaodong Wang, Junbao Zhuo, Shuhao Cui, Shuhui Wang
2021 arXiv   pre-print
To tackle the above issues, we propose Consistency and Diversity Learning (CDL), a simple but effective framework for SSHT by facilitating prediction consistency between two randomly augmented unlabeled  ...  Semi-supervised domain adaptation (SSDA) aims to solve tasks in target domain by utilizing transferable information learned from the available source domain and a few labeled target data.  ...  Method A → C A → P A → R C → A C → P C → R P → A P → C P → R R → A R → C R → P HDA [5] devises a heuristic framework to conduct domain adaptation.  ... 
arXiv:2107.03008v2 fatcat:rse5gdh6unfe5etka5ze3mjmpa

Fair Predictors under Distribution Shift [article]

Harvineet Singh, Rina Singh, Vishwali Mhasawade, Rumi Chunara
2019 arXiv   pre-print
Building on the problem setup of causal domain adaptation, we select a subset of features for training predictors with fairness constraints such that risk with respect to an unseen target data distribution  ...  is minimized.  ...  In the next section, we describe the causal inference framework for the domain adaptation problem, proposed by [Magliacane et al., 2018] , and introduce an identification strategy for the risk minimization  ... 
arXiv:1911.00677v1 fatcat:rwu72mlaybcbzksyn6esxftvom
« Previous Showing results 1 — 15 out of 181,774 results