A Deep Non-linear Feature Mapping for Large-Margin kNN Classification

By Martin Renqiang Min, David A. Stanley, Zineng Yuan, Anthony J. Bonner, Zhaolei Zhang. 2009

97 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Prototypical Networks for Few-shot Learning (2017, methodology): “The DNet-KNN [23] is another margin-based method that improves upon LMNN by utilizing a neural network to perform the embedding instead of a simple linear transformation.”
Online Class-Incremental Continual Learning with Adversarial Shapley Value (2020, methodology): “When we return to Figure 1 and analyze the latent embeddings of memory samples, we observe the natural clustering effect of classes in the embedding space, which has been well-observed previously in the deep learning literature (Min et al. 2009; Donahue et al. 2014).”
Few-Shot Learning Through an Information Retrieval Lens (2017, background): “NCA and LMNN have also been extended to their deep variants [11] and [12], respectively.”
Learning Deep Representation for Imbalanced Classification (2016, background): “However, those classes with high data variability cannot be effectively squeezed to single points [31].”
Adversarial Shapley Value Experience Replay for Task-Free Continual Learning (2020, methodology): “When we return to Figure 1 and analyze the latent embeddings of memory samples, we observe the natural clustering effect of classes in the embedding space, which has been well-observed previously in the deep learning literature [24, 8].”
Training Neural Networks with Stochastic Hessian-Free Optimization (2013, background): “For additional comparison, [29] obtains a mean cl ssification error of 1.”
Class Confidence Weighted kNN Algorithms for Imbalanced Data Sets (2011, methodology): “Experimental results of LMNN show large improvements over conventional k NN and SVM. Min et al. [8] have proposed DNet which uses a non-linear feature mapping method pre-trained with Restricted Boltzmann Machines to achieve the goal of large-margin k NN classi-ﬁcation.”
Deep Supervised t-Distributed Embedding (2010, background): “…construct a generative model of the data (Hinton et al., 2006), learn a classifier (Larochelle et al., 2007), increase the separation between classes (Min et al., 2009; Salakhutdinov & Hinton, 2007; Weston et al., 2008), or preserve pairwise similarities between data points (van der Maaten, 2009).”