Learning Domain-Invariant Subspace Using Domain Features and Independence Maximization
نویسندگان
چکیده
منابع مشابه
Domain-invariant Representation Learning
The learning of domain-invariant representations in the context of domain adaptation with neural networks is considered. We propose a new regularization method that minimizes the domain-specific latent feature representations directly in the hidden activation space. Although some standard distribution matching approaches exist that can be interpreted as the matching of weighted sums of moments,...
متن کاملGraph and Subspace Learning for Domain Adaptation
Graph and Subspace Learning for Domain Adaptation by Le Shu Doctor of Philosophy in Computer and Information Sciences Temple University in Philadelphia, October, 2015 Researcher Advisor: Longin Jan Latecki In many practical problems, given that the instances in the training and test may be drawn from different distributions, traditional supervised learning can not achieve good performance on th...
متن کاملDomain-Invariant Transfer Kernel Learning
Domain invariant learning generalizes a learning model built on training data having different distribution than testing data and still can be accurate. A general principle to tackle this problem is reducing the distribution difference between training data and testing data such that the generalization error can be bounded. Transfer Kernel Learning(TKL) learns the domain invariant kernel by mat...
متن کاملConnecting the Dots with Landmarks: Discriminatively Learning Domain-Invariant Features for Unsupervised Domain Adaptation
Learning domain-invariant features is of vital importance to unsupervised domain adaptation, where classifiers trained on the source domain need to be adapted to a different target domain for which no labeled examples are available. In this paper, we propose a novel approach for learning such features. The central idea is to exploit the existence of landmarks, which are a subset of labeled data...
متن کاملObject Recognition Using Frequency Domain Blur Invariant Features
In this paper, we propose novel blur invariant features for the recognition of objects in images. The features are computed either using the phase-only spectrum or bispectrum of the images and are invariant to centrally symmetric blur, such as linear motion or defocus blur as well as linear illumination changes. The features based on the bispectrum are also invariant to translation, and accordi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Transactions on Cybernetics
سال: 2018
ISSN: 2168-2267,2168-2275
DOI: 10.1109/tcyb.2016.2633306