Transfer Learning on Heterogenous Feature Spaces via Spectral Transformation

By Xiaoxiao Shi, Qi Liu, Wei Fan, Philip S. Yu, Rui Zhu. 2010

179 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, background): “Yan et al. in [257] studied online HDA, where offline labeled data from a source domain is transferred to enhance the online classification performance for the target domain.”
Learning with Augmented Features for Heterogeneous Domain Adaptation (2012, methodology): “For more general HDA tasks, Shi et al. (2010) proposed a method called Heterogeneous Spectral Mapping (HeMap) to discover a common feature subspace by learning two feature mapping matrices as well as the optimal projection of the data from both domains, in which the valuable label information is…”
Hybrid Heterogeneous Transfer Learning through Deep Learning (2014, methodology): “We tune the parameter  for CL-KCCA (see (5) in (Vinokourov, ShaweTaylor, and Cristianini 2002)), mSDA-CCA, and the parameter for HeMap (See (1) in (Shi et al. 2010)) from {0.”
Transfer Neural Trees for Heterogeneous Domain Adaptation (2016, background): “[14] selected a fixed number of source-domain instances to perform heterogeneous spectral mapping (HeMap) for relating cross-domain data.”
Heterogeneous Few-Shot Model Rectification With Semantic Mapping (2020, background): “When the source and target domains have different features, heterogeneous transfer learning [30], [31] extracts structure information or subspaces between two domains [32], [33].”
Rectify Heterogeneous Models with Semantic Mapping (2018, background): “Structure information or subspaces can be found to link two domains (Shi et al., 2010; Wang & Mahadevan, 2011), where sufficient source domain examples should be provided, even the alignment between instances across domains are required (Kulis et al.”
A Survey on Negative Transfer (2020, background): “[95] TBT Feature adaptation D, S Common latent space”
Heterogeneous Domain Adaptation for Multiple Classes (2014, methodology): “[29] proposed a Heterogeneous Spectral Mapping (HeMap) method to learn the mappings based on spectral embedding without using any label information.”
Cross-View Action Recognition over Heterogeneous Feature Spaces (2013, methodology): “[12] proposed a Heterogeneous Spectral Mapping to discover a common feature subspace by learning two feature mapping matrices as well as the optimal projection of the data from both domains.”