Transfer Neural Trees for Heterogeneous Domain Adaptation

Chen, Wei-Yu; Hsu, Tzu-Ming Harry; Tsai, Yao-Hung Hubert; Wang, Yu-Chiang Frank; Chen, Ming-Syan

doi:10.1007/978-3-319-46454-1_25

Cited by 81 publications

(46 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Weakly-shared Deep Transfer Network (DTN) [28] is one of the early attempts in the deep HDA area, which constructs multiple weakly-shared layers in the neural networks. Transfer Neural Trees (TNT) [4] employs a transfer neural decision forest module to adapt important neurons for cross-domain samples. More recently, Yao et al [40] put forward a Soft Transfer Network (STN) approach.…”

Section: Related Workmentioning

confidence: 99%

“…More recently, Heterogeneous Domain Adaptation (HDA) is proposed, where the cross-domain features can be represented by different types of features from various modalities [19,20,38,40,43]. Existing HDA methods can be roughly categorized as supervised [9,14,17,31] or semi-supervised methods [4,18,19,40,42]. Typically, the supervised methods exploit the labeled source domain and a limited number of labeled target domain samples to train the learning algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prototype-Matching Graph Network for Heterogeneous Domain Adaptation

Wang

Luo

Huang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Even though the multimedia data is ubiquitous on the web, the scarcity of the annotated data and variety of data modalities hinder their usage by multimedia applications. Heterogeneous domain adaptation (HDA) has therefore arisen to address such limitations by facilitating the knowledge transfer between heterogeneous domains. Existing HDA methods only focus on aligning the crossdomain feature distributions and ignore the importance of maximizing the margin among different classes, which may lead to a suboptimal classification performance. To tackle this problem, in this paper, we propose the Prototype-Matching Graph Network (PMGN), which gradually explores the domain-invariant class prototype representations. Specifically, we build an end-to-end Graph Prototypical Network, which computes the class prototypes through multiple layers of edge learning, node aggregation, and discrepancy minimization. Our framework utilizes the Swap training strategy to provide adequate supervision for training the edge learning component. Moreover, the proposed PMGN can be equipped with the clustering module that utilises the KL-divergence as a distance metric to reduce the distribution difference between the source and target data. Extensive experiments on three HDA tasks (i.e. object recognition, text-to-image classification, and text categorization) demonstrate the superiority of our approach over the state-of-theart HDA methods. CCS CONCEPTS • Computing methodologies → Transfer learning.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Prototype-Matching Graph Network for Heterogeneous Domain Adaptation

Wang

Luo

Huang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…Chen et al [65] proposed a neural network based framework called Transfer Neural Trees (TNT) for semi-supervised HDA tasks. This framework is divided into two layers: mapping and prediction.…”

Section: Tntmentioning

confidence: 99%

“…For adaptation and classification, Chen et al [65] also proposed Transfer Neural Decision Forest (Transfer-NDF) for use in the TNT framework. Inspired from deep neural decision forest [66] and random forest [42], Transfer-NDF uses neural networks as decision trees as to build a forest of neural decision trees.…”

Section: Tntmentioning

confidence: 99%