Learning semi-supervised nonlinear embeddings for domain-adaptive pattern recognition

We study the problem of learning nonlinear data embeddings in order to obtain representations for efficient and domain-invariant recognition of visual patterns. Given observations of a training set of patterns from different classes in two different domains, we propose a method to learn a nonlinear mapping of the data samples from different domains into a common domain. The nonlinear mapping is learnt such that the class means of different domains are mapped to nearby points in the common domain in order to appropriately align the two domains. Meanwhile, unlabeled samples are also used in the computation of the embedding via an objective term representing the preservation of the global geometry of the data. Along with the mapping of the training points, we also learn a linear classifier in the common domain, which allows an accurate estimation of the unknown class labels. We evaluate the performance of the proposed algorithm in domain-adaptive face and object recognition experiments. Experimental results show that the proposed method yields quite promising performance, outperforming baseline domain adaptation methods.
International Conference on Advanced Technologies, Computer Engineering and Science (ICATCES 2019), Apr 26-28, 2019


Learning Smooth Pattern Transformation Manifolds
Vural, Elif (2013-04-01)
Manifold models provide low-dimensional representations that are useful for processing and analyzing data in a transformation-invariant way. In this paper, we study the problem of learning smooth pattern transformation manifolds from image sets that represent observations of geometrically transformed signals. To construct a manifold, we build a representative pattern whose transformations accurately fit various input images. We examine two objectives of the manifold-building problem, namely, approximation a...
Out-of-Sample Generalizations for Supervised Manifold Learning for Classification
Vural, Elif (2016-03-01)
Supervised manifold learning methods for data classification map high-dimensional data samples to a lower dimensional domain in a structure-preserving way while increasing the separation between different classes. Most manifold learning methods compute the embedding only of the initially available data; however, the generalization of the embedding to novel points, i.e., the out-of-sample extension problem, becomes especially important in classification applications. In this paper, we propose a semi-supervis...
Learning Time-Vertex Dictionaries for Estimating Time-Varying Graph Signals
Acar, Abdullah Burak; Vural, Elif (2022-01-01)
In this work, we study the problem of learning time-vertex dictionaries for the modeling and estimation of time-varying graph signals. We consider a setting with a collection of partially observed time-varying graph signals, and propose a solution for the estimation of the missing signal observations by learning time-vertex dictionaries from the available observations. We adopt a time-vertex dictionary model defined through a set of joint time-vertex spectral kernels, each of which captures a different spec...
Approximation of pattern transformation manifolds with parametric dictionaries
Vural, Elif (2011-07-12)
The construction of low-dimensional models explaining high-dimensional signal observations provides concise and efficient data representations. In this paper, we focus on pattern transformation manifold models generated by in-plane geometric transformations of 2D visual patterns. We propose a method for computing a manifold by building a representative pattern such that its transformation manifold accurately fits a set of given observations. We present a solution for the progressive construction of the repr...
Optimising a nonlinear utility function in multi-objective integer programming
Ozlen, Melih; Azizoğlu, Meral; Burton, Benjamin A. (2013-05-01)
In this paper we develop an algorithm to optimise a nonlinear utility function of multiple objectives over the integer efficient set. Our approach is based on identifying and updating bounds on the individual objectives as well as the optimal utility value. This is done using already known solutions, linear programming relaxations, utility function inversion, and integer programming. We develop a general optimisation algorithm for use with k objectives, and we illustrate our approach using a tri-objective i...
Citation Formats
E. Vural, “Learning semi-supervised nonlinear embeddings for domain-adaptive pattern recognition,” Alanya, Turkey, 2019, p. 295, Accessed: 00, 2021. [Online]. Available: http://muh.karabuk.edu.tr/bilgisayar/icatces/proceeding_book_2019.pdf.