Saved in:
Bibliographic Details
Main Authors: Wang, Shixiong, Wang, Haowei, Li, Xinke, Honorio, Jean
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2301.13565
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908344780849152
author Wang, Shixiong
Wang, Haowei
Li, Xinke
Honorio, Jean
author_facet Wang, Shixiong
Wang, Haowei
Li, Xinke
Honorio, Jean
contents Trustworthy machine learning aims at combating distributional uncertainties in training data distributions compared to population distributions. Typical treatment frameworks include the Bayesian approach, (min-max) distributionally robust optimization (DRO), and regularization. However, three issues have to be raised: 1) the prior distribution in the Bayesian method and the regularizer in the regularization method are difficult to specify; 2) the DRO method tends to be overly conservative; 3) all the three methods are biased estimators of the true optimal cost. This paper studies a new framework that unifies the three approaches and addresses the three challenges above. The asymptotic properties (e.g., consistencies and asymptotic normalities), non-asymptotic properties (e.g., generalization bounds and unbiasedness), and solution methods of the proposed model are studied. The new model reveals the trade-off between the robustness to the unseen data and the specificity to the training data. Experiments on various real-world tasks validate the superiority of the proposed learning framework.
format Preprint
id arxiv_https___arxiv_org_abs_2301_13565
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Learning Against Distributional Uncertainty: On the Trade-off Between Robustness and Specificity
Wang, Shixiong
Wang, Haowei
Li, Xinke
Honorio, Jean
Machine Learning
62F35, 62G35, 62C12
Trustworthy machine learning aims at combating distributional uncertainties in training data distributions compared to population distributions. Typical treatment frameworks include the Bayesian approach, (min-max) distributionally robust optimization (DRO), and regularization. However, three issues have to be raised: 1) the prior distribution in the Bayesian method and the regularizer in the regularization method are difficult to specify; 2) the DRO method tends to be overly conservative; 3) all the three methods are biased estimators of the true optimal cost. This paper studies a new framework that unifies the three approaches and addresses the three challenges above. The asymptotic properties (e.g., consistencies and asymptotic normalities), non-asymptotic properties (e.g., generalization bounds and unbiasedness), and solution methods of the proposed model are studied. The new model reveals the trade-off between the robustness to the unseen data and the specificity to the training data. Experiments on various real-world tasks validate the superiority of the proposed learning framework.
title Learning Against Distributional Uncertainty: On the Trade-off Between Robustness and Specificity
topic Machine Learning
62F35, 62G35, 62C12
url https://arxiv.org/abs/2301.13565