Unsupervised domain adaptation with copula models

@article{Tran2017UnsupervisedDA,
  title={Unsupervised domain adaptation with copula models},
  author={Cuong D. Tran and Ognjen Rudovic and Vladimir Pavlovic},
  journal={2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP)},
  year={2017},
  pages={1-6}
}
We study the task of unsupervised domain adaptation, where no labeled data from the target domain is provided during training time. To deal with the potential discrepancy between the source and target distributions, both in features and labels, we exploit a copula-based regression framework. The benefits of this approach are two-fold: (a) it allows us to model a broader range of conditional predictive densities beyond the common exponential family; (b) we show how to leverage Sklar's theorem… 

Tables from this paper

Domain Adaptation on Graphs by Learning Aligned Graph Bases

This paper proposes a method for domain adaptation on graphs based on learning the spectrum of the label function in a source graph with many labels, and transferring the spectrum information to the target graph through a transformation between the Fourier bases of the two graphs that flexibly “aligns” them.

Graph Domain Adaptation with Localized Graph Signal Representations

A domain adaptation algorithm designed for graph domains that estimates the target labels of unknown target labels by making use of the similarity between the characteristics of the variation of the label functions on the two graphs.

References

SHOWING 1-10 OF 23 REFERENCES

Domain Adaptation via Transfer Component Analysis

This work proposes a novel dimensionality reduction framework for reducing the distance between domains in a latent space for domain adaptation and proposes both unsupervised and semisupervised feature extraction approaches, which can dramatically reduce thedistance between domain distributions by projecting data onto the learned transfer components.

Return of Frustratingly Easy Domain Adaptation

This work proposes a simple, effective, and efficient method for unsupervised domain adaptation called CORrelation ALignment (CORAL), which minimizes domain shift by aligning the second-order statistics of source and target distributions, without requiring any target labels.

Unsupervised Domain Adaptation by Domain Invariant Projection

This paper learns a projection of the data to a low-dimensional latent space where the distance between the empirical distributions of the source and target examples is minimized and demonstrates the effectiveness of the approach on the task of visual object recognition.

Semi-Supervised Domain Adaptation with Non-Parametric Copulas

A new framework based on the theory of copulas is proposed to address semi-supervised domain adaptation problems, and a novel vine copula model is introduced, which allows for this factorization in a non-parametric manner.

Learning an Invariant Hilbert Space for Domain Adaptation

This paper introduces a learning scheme to construct a Hilbert space (i.e., a vector space along its inner product) to address both unsupervised and semi-supervised domain adaptation problems to simultaneously minimizing a notion of domain variance while maximizing a measure of discriminatory power.

Frustratingly Easy Semi-Supervised Domain Adaptation

This work builds on the notion of augmented space and harnesses unlabeled data in target domain to ameliorate the transfer of information from source to target and can be applied as a pre-processing step to any supervised learner.

Robust speech recognition using multivariate copula models

This work extends this work from classification to recognition and elaborate further on the mathematical properties of the copula models in the form of lemmas and shows that the proposed copula-based models improve the accuracy over a comparable baseline.

A Semiparametric Gaussian Copula Regression Model for Predicting Financial Risks from Earnings Calls

This work quantitatively study how earnings calls are correlated with the financial risks, with a special focus on the financial crisis of 2009, and proposes the use of copula: a powerful statistical framework that separately models the uniform marginals and their complex multivariate stochastic dependencies.

Copula Estimation

A brief survey of estimation procedures for copula models, which reviews parametric, semiparametric and nonparametric approaches to inference on copulas for random samples with dependent marginals and copula-based time series and also discusses several problems of robust estimation in these frameworks.