DOMAIN ADAPTATION VIA TRANSFERRING SPECTRAL PROPERTIES OF LABEL FUNCTIONS ON GRAPHS

2016-07-12
Pilanci, Mehmet
Vural, Elif
We propose a domain adaptation algorithm that relies on a graph representation of data samples in the source and target domains. The algorithm combines the information of the known class labels in the source and target domains through the Fourier coefficients of the class label function in the two graphs. The proposed method does not require an ordering or a one-to-one mapping between the samples of the source and target domains; instead, it uses only a small set of matched pairs that serve the purpose of "aligning" the source and target Fourier bases. The estimation of the coefficients of the label function in the source and target Fourier bases is then formulated as a simple convex optimization problem. The proposed domain adaptation algorithm is tested in face recognition under varying pose and illumination and is observed to provide significant improvement over reference classification approaches especially when the data distributions in the source and target domains differ significantly.
12th IEEE Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)

Suggestions

Domain Adaptation on Graphs via Frequency Analysis
Pilancı, Mehmet; Vural, Elif (2019-08-22)
Classical machine learning algorithms assume the training and test data to be sampled from the same distribution, while this assumption may be violated in practice. Domain adaptation methods aim to exploit the information available in a source domain in order to improve the performance of classification in a target domain. In this work, we focus on the problem of domain adaptation in graph settings. We consider a source graph with many labeled nodes and aim to estimate the class labels on a target graph wit...
Domain Adaptation with Nonparametric Projections
Vural, Elif (2019-08-22)
Domain adaptation algorithms focus on a setting where the training and test data are sampled from related but different distributions. Various domain adaptation methods aim to align the source and target domains in a new common domain by learning a transformation or projection. In this work, we learn a nonlinear and nonparametric projection of the source and target domains into a common domain along with a linear classifier in the new domain. Experiments on image data sets show that the proposed nonlinear a...
Domain adaptation on graphs by learning aligned graph bases
Pilancı, Mehmet; Vural, Elif; Department of Electrical and Electronics Engineering (2018)
In this thesis, the domain adaptation problem is studied and a method for domain adaptation on graphs is proposed. Given sufficiently many observations of the label function on a source graph, we study the problem of transferring the label information from the source graph to a target graph for estimating the target label function. Our assumption about the relation between the two domains is that the frequency content of the label function, regarded as a graph signal, has similar characteristics over the so...
Estimation of partially observed multiple graph signals by learning spectrally concentrated graph kernels
Turhan, Gülce; Vural, Elif; Department of Electrical and Electronics Engineering (2021-3-31)
Graph models provide flexible tools for the representation and analysis of signals defined over domains such as social or sensor networks. However, in real applications data observations are often not available over the whole graph, due to practical problems such as broken sensors, connection loss, or storage problems. In this thesis, we study the problem of estimating partially observed graph signals on multiple graphs. We consider possibly multiple graph domains over which a set of signals is available wi...
Multi-resolution visualization of large scale protein networks enriched with gene ontology annotations
Yaşar, Sevgi; Can, Tolga; Department of Computer Engineering (2009)
Genome scale protein-protein interactions (PPIs) are interpreted as networks or graphs with thousands of nodes from the perspective of computer science. PPI networks represent various types of possible interactions among proteins or genes of a genome. PPI data is vital in protein function prediction since functions of the cells are performed by groups of proteins interacting with each other and main complexes of the cell are made of proteins interacting with each other. Recent increase in protein interactio...
Citation Formats
M. Pilanci and E. Vural, “DOMAIN ADAPTATION VIA TRANSFERRING SPECTRAL PROPERTIES OF LABEL FUNCTIONS ON GRAPHS,” presented at the 12th IEEE Image, Video, and Multidimensional Signal Processing Workshop (IVMSP), Bordeaux, FRANCE, 2016, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/52878.