Recognizing heterogeneous cross-domain data via generalized joint distribution adaptation
| Autor: | Yao-Hung Hubert Tsai, Yu-Chiang Frank Wang, Shi-Yen Tao, Yuan-Ting Hsieh, Yi-Ren Yeh |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Matching (graph theory)
Computer science business.industry Pattern recognition 02 engineering and technology Manifold Domain (software engineering) Joint probability distribution Feature (computer vision) 020204 information systems 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence Projection (set theory) business Subspace topology Curse of dimensionality Sparse matrix |
| Zdroj: | ICME |
| DOI: | 10.1109/icme.2016.7552878 |
| Popis: | In this paper, we propose a novel algorithm of Generalized Joint Distribution Adaptation (G-JDA) for heterogeneous domain adaptation (HDA), which associates and recognizes cross-domain data observed in different feature spaces (and thus with different dimensionality). With the objective to derive a domain-invariant feature subspace for relating source and target-domain data, our G-JDA learns a pair of feature projection matrices (one for each domain), which allows us to eliminate the difference between projected cross-domain heterogeneous data by matching their marginal and class-conditional distributions. We conduct experiments on cross-domain classification tasks using data across different features, datasets, and modalities. We confirm that our G-JDA would perform favorably against state-of-the-art HDA approaches. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|