Local Fisher discriminant analysis for supervised dimensionality reduction

By Masashi Sugiyama. International Conference on Machine Learning, 2006

409 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/62204f0cf9d70503cb42a73c198504e412137dd0

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. Tao A. Xiang
2. Masashi Sugiyama
3. Jiliang Tang
4. Shaogang Gong
5. Qi Tian
6. Jundong Li
7. K. Müller
8. Huan Liu

In Computer Vision and Pattern Recognition

1. Yi Yang
2. Qi Tian
3. Tao A. Xiang
4. Shaogang Gong
5. Liang Zheng
6. Xiaogang Wang
7. Dacheng Tao
8. Chunhua Shen

Citations in the corpus

• Local Fisher Discriminant Analysis for Pedestrian Re-identification (2013, methodology): “This ob- servation suggests that PCA encoding is crucial to remove the noise that may reduce the richness of the LFDA learnt subspace.”
• Text Classification Algorithms: A Survey (2019, methodology): “From the fisher discriminant analysis (FDA) [76,78], we can re-write the equation as:”
• How to solve classification and regression problems on high-dimensional data with a supervised extension of slow feature analysis (2013, background): “Sugiyama (2006) proposed local FDA (LFDA), an adaptation of FDA with a discriminative objective function that also preserves the local structure of the input data.”
• Feature Selection (2016, background): “, 2011) propose to utilize the local discriminative information (Sugiyama, 2006; Yang et al., 2010) to select discriminative features.”
• Learning a Discriminative Null Space for Person Re-identification (2016, background): “Among them, the distance metric learning based methods are RPLM [15], MtMCML [27], Mid-level Filter [41], SCNCD [38], Similarity Learning [3], LADF [21], ITML [5], LMNN [35], KISSME [17], and MCML [9], whilst the others are discriminative subspace learning based methods including kCCA [24], MFA [36]”
• Covariate Shift Adaptation by Importance Weighted Cross Validation (2007, methodology): “For linear regression, IWLS-based active learning methods that focus on minimizing the variance of the estimator have been developed (Wiens, 2000; Sugiyama, 2006a).”
• Enhancing Person Re-identification in a Self-Trained Subspace (2017, methodology): “[50] is the semi-supervised extension of local fisher discriminant analysis (LFDA) [52] which combines the supervised LFDA and the unsupervised PCA to jointly exploit labeled and unlabeled data.”
• Person Re-Identification Using Kernel-Based Metric Learning Methods (2014, methodology): “To avoid this problem, we propose to use a kernel approach, based on the method introduced in [22] in the context of supervised dimensionality reduction.”