Transfer Hashing with Privileged Information

By Joey Tianyi Zhou, Xinxing Xu, Sinno Jialin Pan, Ivor Wai-Hung Tsang, Zheng Qin, Rick Siow Mong Goh. 2016

29 citations recorded by the source corpus.

Most existing learning to hash methods assume that there are sufficient data, either labeled or unlabeled, on the domain of interest (i.e., the target domain) for training. However, this assumption cannot be satisfied in some real-world applications. To address this data sparsity issue in hashing, inspired by transfer learning, we propose a new framework named Transfer Hashing with Privileged Information (THPI). Specifically, we extend the standard learning to hash method, Iterative Quantization (ITQ), in a transfer learning manner, namely ITQ+. In ITQ+, a new slack function is learned from auxiliary data to approximate the quantization error in ITQ. We developed an alternating optimization approach to solve the resultant optimization problem for ITQ+. We further extend ITQ+ to LapITQ+ by utilizing the geometry structure among the auxiliary data for learning more precise binary codes in the target domain. Extensive experiments on several benchmark datasets verify the effectiveness of our proposed approaches through comparisons with several state-of-the-art baselines.

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

1. Yong Liu
2. Yong Kiam Tan
3. Chen Gong
4. Xiaojun Chang
5. Shijian Lu
6. Meng Fang
7. Xinxing Xu
8. Romain Vial

In Computer Vision and Pattern Recognition

1. Yong Liu
2. Yong Kiam Tan
3. Chen Gong
4. Xiaojun Chang
5. Shijian Lu
6. Meng Fang
7. Xinxing Xu
8. Romain Vial

Citations in the corpus

• Improved Recurrent Neural Networks for Session-based Recommendations (2016, methodology): “ normalization [12] and residual connections [9]. We seek to apply some of these methods to enhance the training of our recommendation RNNs. The learning using privileged information (LUPI) framework [27, 30, 32] was proposed to utilize the additional feature representations that are only available ”
• Robust Unsupervised Cross-modal Hashing for Multimedia Retrieval (2020, unknown): “Without loss of generality, we assume that all objects are zero-centered, which is usually adopted in existing hashing algorithms [36, 46].”
• TORNADO: A Spatio-Temporal Convolutional Regression Network for Video Action Proposal (2017, background): “Developing a fast and accurate algorithm which can well mimic this human capability can unlock great potentials for a series of responsive and automatic applications, such as assistive robots and video surveillance [7, 43, 24, 20].”
• Teaching Semi-Supervised Classifier via Generalized Distillation (2018, methodology): “LUPI has been widely used in many machine learning scenarios such as metric learning [Yang et al. , 2016], transfer hashing [Zhou et al. , 2016], multi-label learning [Y-ou et al. , 2017], and deep learning [Hu et al. , 2016].”