Stochastic Learning on Imbalanced Data: Determinantal Point Processes for Mini-batch Diversification

By Cheng Zhang, Hedvig Kjellström, Stephan Mandt. 2017

58 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Scalable Recollections for Continual Lifelong Learning (2017, background): “One approach to achieving this is a Determinantal Point Process (DPP) (Kulesza et al., 2012) as recently proposed for selecting diverse neural network mini-batches (Zhang et al., 2017).”
Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds (2019, methodology): “[39] Cheng Zhang, Hedvig Kjellstrom, and Stephan Mandt.”
Gradient Diversity Empowers Distributed Learning (2017, methodology): “[48] propose a non-uniform sampling scheme that can increase the chance of getting more diverse data in a batch.”
Gradient Diversity: a Key Ingredient for Scalable Distributed Learning (2017, methodology): “De et al. (2016) propose an optimization algorithm for choosing the batch-size, and weighted sampling techniques have also been developed (Needell and Ward, 2016, Zhang et al., 2017, Zhao and Zhang, 2014).”
Differentiable Scaffolding Tree for Molecular Optimization (2021, unknown): “DPP models the repulsive correlation between data points [28] and has been successfully applied to many applications such as text summarization [7], mini-batch sampling [50], and recommendation system [6].”
The Effect of Network Width on the Performance of Large-batch Training (2018, background): “Gradient Diversity Previous work indicates that mini-batch can achieve better convergence rates by increasing the diversity of gradient batches, e .g., using stratiﬁed sampling [38], Determinantal Point Processes [39], or active sampling [40].”
Generative Knowledge Distillation for General Purpose Function Compression (2017, background): “One approach to achieving this is a Determinantal Point Process (DPP) [14] as recently proposed for selecting diverse neural network mini-batches [32].”
Advances in Variational Inference (2017, background): “stratified sampling [225], where one samples data from predefined subgroups based on meta-data or labels, clusteringbased sampling [47], which amounts to clustering the data using k-means and then sampling data from every cluster with adjusted probabilities, and diversified mini-batch sampling [223]”