Why resampling outperforms reweighting for correcting sampling bias with stochastic gradients
A data set sampled from a certain population is biased if the subgroups of the population are sampled at proportions that are significantly different from their underlying proportions. Training machine learning models on biased data sets requires correction techniques to compensate for the bias. We consider two commonly-used techniques, resampling and reweighting, that rebalance the proportions of the subgroups to maintain the desired objective function. Though statistically equivalent, it has been observed that resampling outperforms reweighting when combined with stochastic gradient algorithms. By analyzing illustrative examples, we explain the reason behind this phenomenon using tools from dynamical stability and stochastic asymptotics. We also present experiments from regression, classification, and off-policy prediction to demonstrate that this is a general phenomenon. We argue that it is imperative to consider the objective function design and the optimization algorithm together while addressing the sampling bias.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Importance Resampling for Off-policy Prediction
Importance sampling (IS) is a common reweighting strategy for off-policy prediction in reinforcement learning. While it is consistent and unbiased, it can result in high variance updates to the weights for the value func…
PredictionReinforcement LearningImportance Resampling for Off-policy Policy Evaluation
Importance sampling is a common approach to off-policy learning in reinforcement learning. While it is consistent and unbiased, it can result in high variance updates to the parameters for the value function. Weighted i…
Dirichlet-based Per-Sample Weighting by Transition Matrix for Noisy Label Learning
For learning with noisy labels, the transition matrix, which explicitly models the relation between noisy label distribution and clean label distribution, has been utilized to achieve the statistical consistency of eithe…
Learning with noisy labelsWhen resampling/reweighting improves feature learning in imbalanced classification?: A toy-model study
A toy model of binary classification is studied with the aim of clarifying the class-wise resampling/reweighting effect on the feature learning performance under the presence of class imbalance. In the analysis, a high-d…
Binary Classificationimbalanced classificationUBR$^2$S: Uncertainty-Based Resampling and Reweighting Strategy for Unsupervised Domain Adaptation
Unsupervised domain adaptation (UDA) deals with the adaptation process of a model to an unlabeled target domain while annotated data is only available for a given source domain. This poses a challenging task, as the doma…
Domain AdaptationUnsupervised Domain Adaptation