Domain adaptation assumes that samples from source and target domains are freely accessible during a training phase. However, such an assumption is rarely plausible in the real world possibly causes data privacy issues, especially when label of domain can be sensitive attribute as identifier. To avoid accessing could contain information, we introduce free (SFDA). Our key idea to leverage pretra...

Photoacoustic tomography (PAT) is a novel hybrid medical imaging technique that aims to combine the large contrast of optical coefficients with the high resolution capabilities of ultrasounds. We assume that the first step of PAT, namely the reconstruction of a map of absorbed radiation from boundary ultrasound measurement, is done. We focus on quantitative photoacoustic tomography (QPAT), whic...

To attain the benefits of simultaneous shooting in marine data acquisition, either the simultaneous sources have to be placed far apart to minimize the interferences from one another or sufficient attenuation of the interferences is necessary. Attenuating the interferences in a domain where they are randomly distributed has a potential of attenuating the signals particularly when they have conf...

Feature selection is an effective approach to reducing dimensionality by selecting relevant original features. In this work, we studied a novel problem of multi-source feature selection for unlabeled data: given multiple heterogeneous data sources (or data sets), select features from one source of interest by integrating information from various data sources. In essence, we investigate how we c...